Transferrin and transferrin receptors in normal brain and in ...

1 downloads 0 Views 386KB Size Report
(I) or I O O ~ M (2 and 3) "51-labelled N'-ASA Kallidin. Tracks. 44; labelling of live cells (10' cells/ml) in serum-free medium for 60min at 37°C in the presence (6) ...
89 1

622nd M E E T I N G , LEICESTER A

-1

2

B

4

3

5

6

7

8

zoo b

92 rn

69

*

46

Fig. I . Labelling

bearing wheat germ agglutinin. After extensive washing, specific elution of bound material was achieved by incubating the washed beads with minimal volumes of 0 . 5 ~ - N acetyl glucosamine in 0.5% deoxycholate. Eluted material ran with the same mobility (Fig. 1B, tracks 7 and 8) as that directly solubilized from labelled membranes. When live cells were incubated with neuraminidase before labelling, there was a distinct shift to higher mobility (Fig. IB, track 6) of the labelled band, which suggests the presence of sialic acid residues in the native form. We conclude that the W-ASA kallidin probe binds to physiologically relevant bradykinin receptors on the NG108-15 cell membrane, and that the receptor is a sialidated glycoprotein with a molecular mass of 166 000 Da. The photoaffinity probe will be useful in examining bradykinin receptors o n other tissues and its use may facilitate the isolation of the bradykinin receptor. Hamprecht. B. (1977) lnr. R P V .C,v/ol. 49. 99-170 Higashida, H . & Brown, D. A . (19x6) Nuture ( L ~ n d ~323, n ) 333-335 Higashida. H.. Streaty. R . . Klee. W. & Nirenberg, M . (1986) Proc. N u / / . Acud. Sci. U . S . A . 83, 942 946 Hunter. W. M . & Greenwood, F. C . (1962) Nuture (London) 194.

495 496 bradykinin receptors on NG 108-15 cells Laemmli, U . K. (1970) Na/ure (London) 227. 6x0 685 with N - A S A Kullidin Reiser, G . & Hamprecht. B. (1982) Bruin Res. 239, 191 199 Tracks 1-3: tumour membranes were photolytically labelled in Reiser. G. & Hamprecht, B. (1985) Pflurgrrs Arch. 405, 26C264 the absence (1.3) or presence (2) or IOp-bradykinin and 4 0 p ~ Reiser. G . . Walter. U. & Hamprecht. B. (1984) Bruin Res. 290, 367 371 ( I ) or I O O ~ M(2 and 3) "51-labelled N'-ASA Kallidin. Tracks 4 4 ; labelling of live cells (10' cells/ml) in serum-free medium Yano. K.. Hagashida. H.. Inoue, R. & Nozawa, Y . (1984) J . B i d . Chem. 259. 10201 10207 for 60min at 37°C in the presence (6) or absence (4 and 5 ) of ncuraminidase ( 5 u./ml), then labelled in the presence (4) o r absence (5 and 6) of IOpM-bradykinin. Tracks 7 and 8; tumour membranes were labelled in the absence (7) and presence (8) of 10 pwbradykinin, solubilized, absorbed and eluted from a wheatgerm agglutinin column, as described in the text. Arrows indicate the positions of molecular mass markers at molecular mass 200, 92, 69 and 46 kDa. Received 20 March 1987 of

Transferrin and transferrin receptors in normal brain and in Alzheimer's disease C H R I S T O P H E R M . MORRIS*, J E N N I F E R A. C O U R T * , ALI A. MOSHTAGHIES, A N D R E W SKILLENS, J O H N M . C A N D Y , R O B E R T H . PERRY?, J A M E S A. E D W A R D S O N * and A N D R E W FAIRBAIRN$ * M .R.C. N~Juropndocrinolo~y Unit und t D q u r tmen t of Pu th o logv , Ne w~*u.s t le General Hospital, Newcastle upon Tyne NE4 6BE, 1Depurtmmt of Clinical Biochemistry. University of' Nenwstlc upon Tvne NE2 4 H H , and $St Nicholus Hospitul Go.sfi)rth, NE3 3 X T . U.K. The abnormal accumulation of neurotoxic metal ions may play a role in the pathogenesis of a number of neurodegenerative disorders. for example aluminium is associated with the major neuropathological features of Alzheimer's disease (Per1 & Brody. 1980; Candy pt al., 1986). The iron transport protein, transferrin has a high affinity for aluminium (Cochran ct ul., 1984) and transferrin receptors present o n the lumen of brain capillaries (Jefferies ct a/., 1984). may mediate its uptake into brain (J. M . Candy, J . A. Edwardson. R. Faircloth, A. B. Keith, C. M . Morris & R. G . L. Pullen. unpublished work). Furthermore, the aluminium complex of transferrin has been shown t o be as efficiently bound and internalized by cultured human neuroAbbreviation used: SDAT, senile dementia of the Alzheimer type; Fe Tf. Fc transferrin.

Vol. 15

blastoma I M R 32 cells as the iron transferrin complex (Morris r t ul., 1987). Transferrin may play a n important role in the transport of several metal ions in brain as relatively high levels of transferrin m R N A are present in brain (Levin et d.,1984). In the present study, the levels of transferrin and transferrin receptors have been determined in the cerebral cortex of cases of senile dementia of the Alzheimer type (SDAT) and in elderly, mentally normal patients (Table 1) (for diagnostic criteria see Tomlinson et al., 1968, 1970). Coronal brain slices were rapidly frozen immediately after autopsy in melting freon (Arcton 12, ICI) and stored at -70°C. Frontal (Brodmann areas 8/9) and temporal cortex (Brodmann areas 2 1/22) were rapidly subdissected into white and grey matter after thawing, homogeized a t 4°C in 2 IOvol. phosphate buffered saline, p H 7.4 and centrifuged for 5min at I O O O O g and 4°C. The supernatants and pellets were carefully separated and transferrin levels were determined in the supernatants by nephelometry, using a modification of the method of Spencer & Price (1979) and a Cobas Bio Centrifugal Fast Analyser. Transferrin receptors were studied in grey-matter membranes using '251-labelled Fe-transferrin ('-'I-labelled Fe-Tf). The iron complex of transferrin was prepared according to Klausner r t ul. (1983) and the complex iodinated with chloramine-T (Karin & Mintz, 1981). The cortical membranes were washed three times with ice-cold 50 m w s o d i u m phosphate buffer, p H 7.4

892

BIOCHEMICAL SOCl ETY TRANSACTIONS

and re-suspended in phosphate buffer containing albumin 1 mg/ml at a dilution of approximately lOmg membrane proteiniml. For single-point binding assays, the membrane suspensions were incubated with I nM-”’l-labeIled Fe--Tf (specific activity 1 kBqifmol) and for Scatchard analysis with concentrations of ‘”I-labelled Fe- Tf ranging from 0.125-- 16 nM. The total assay volume was 250p1 and the incubations. for I h at 4°C. were terminated by centrifugation at I00OOg and 4°C for 3min. After aspiration of supernantant. the pellets were counted. Non-specific binding, determined in the presence of 1 pM-Fe-Tfranged between 20 and 45% of the total bound ligand. The binding of ‘”I-labelled Fe-Tf to receptors was also determined using tritium film autoradiography. Cryostat sections (20 pm) of unfixed frontal and temporal cortex, from five SDAT cases and five elderly normal controls were thaw mounted on to gelatine-coated slides. Sections were washed for 20min at 4°C in phosphate buffer (as above), then incubated with 1 n~-”’l-labelled Fe-Tf for 1 h at 4°C. Sections were washed three times for 2min. rapidly air dried and exposed to Ultrofilm (LKB) for 48h. Non-specific binding, which was determined in the presence of 1 pM-Fe-Tf, was negligible under these conditions. A Magiscan 2A image-analysis system (Joyce Loebel) was used to quantify binding. Table I . Transfcrrin l i w h and transfivrin binding in the.fiontal tortes o/ S D A T and cmtrol cases Results arc means S.D. SDAT Transferrin levels: (pg/mg protein) grey matter white matter Transferin binding (fmol mg protein)

I.X3

1.95

k 1.70 k 1.24

11.8 i 7 . 1

6

I1

Age (years) Male, female Post-mortem delay ( h )

xo f 5 2;4 25 k 14

Elderlv controls

2.12 f 0.66 1.76 f 0.58 12.8 f 3.6 6 80 f 4 2,4 24 f 12

The levels of transferrin in frontal cortex were similar in the SDAT cases and normal patients (Table I ) . Transferrin levels were also similar in the temporal cortex of the SDAT and normal patients (e.g. values in grey matter 2.63 f 1.70 and 2.66 f 2.1 1 pg/mg protein, respectively). I n addition. the density of binding sites was not significantly different in SDAT and elderly controls as demonstrated by both receptor and autoradiography and binding to cortical membranes (Table I ) . Scatchard analysis of pooled membrane samples indicated the presence of a single high-affinity binding site for transferrin in both the SDAT and control cases, with K,s ranging from 2-7nM and maximal binding of 2C30 fmol/mg protein. These results show the presence of high levels of transferrin in human cerebral cortex that cannot be accounted for by cerebral blood content. A single high-affinity site for transferrin is present in the cerebral cortex with a Kd similar to that reported in cultured cells (Morris ot d.. 1987). The level of transferrin and density of transferrin-binding sites is unchanged in SDAT, suggesting that the transport of iron and aluminium is not impaired in SDAT. Candy. J . M.. Oakley. A. E., Klinowski. J.. Carpenter. T. A., Perry. R . 11.. Atack. J . R.. Perry. E. K.. Blessed. G.. Fairbairn. A . & Edwardson. J. A . (19x6) Ltrnc.c,/ i. 154 357 Cochran. M.. Coates. J . & Neoh. S . (1984) F E B S I , ( , / / . 176. 129 132 JelTeries. W . A,. Brandon. M . R.. €lunt. S. V.. Williams. A . F., Gatter, K. C. & David. D. Y . (19x4) N o / u r c ~ ( L o i i [ l o i312. i) I62 163 Karin. M . & Mintz. B. (1981) J . B i d . C/UVJJ. 256. 3245 3252 Klausncr. R . D.. Van Renswoude. J.. Ashwell. G., Kcinpf. C.. Schcchter. A. N., Dean, A. & Bridges, K . R. (1983) J . B i d . C/ICW. 258. 4715 4724 Levin. M . J., Tuil, D.. IJzan, G., Drcyfus, J.-C. & Kahn. A . (19x4) B I ( I ~ ~ / ILiiopliJ.\. (vJI. Kc.v. C’oniniun. 122, 2 I2 2 I 7 Morris. C. M.. Candy, J. M.. Court. J . A,. Whitford. C. A. & Edwardson. J. A. (1987) Bioc~/icw.S o t . . Trtrns. in the press Perl. D. P. & Brody. A. R . (1980) St~icwcc~ 208. 297 209 Spencer. K . & Price. C. P. (1979) c‘lin. Cliinr. At,/cr 95. 263 276 Tomlinson. B. E., Blessed. G . & Roth. M . (196X) J . Ncirrol. SCI.7. 331 356 Tomlinson, B. E.. Blessed. G. & Roth. M . (1970) J . Ncurol. Sci. 11. 205 242 Received 26 March 19x7

Isolation of a cytosolic protein (P36) from pig brain by benzodiazepine-affinity chromatography EWEN F. KIRKNESS* and ANTHONY J. TURNER Dcpurtnicnt yf Biochmiistry, University of Leeds. L t w h LS2 9JT. U . K . The benzodiazepines are a major class of drugs, used principally for the treatment of anxiety, but also possessing sedative, muscle-relaxant and anti-convulsant properties. They are thought to mediate their principal pharmacological actions through binding to a ‘central’ receptor which comprises part of a 7-aminobutyrate-regulated anion channel (Tallman ot al.. 1980). A number of these drugs are, however, also capable of binding to distinct membrane-bound proteins referred to as ‘peripheral’ and ‘micromolar’ receptors, and of modulating the activity of soluble enzymes, e.g. brain aldehyde reductase (Braestrup & Squires, 1977; Bowling & DeLorenzo, 1982; Javors & Erwin, 1980). Although specific pharmacological consequences of these interactions remain to be established. the binding of benzodiazepines to tissue proteins appears to result in their accumulation in the body and to their low rate of elimination (Guentert, 1984). Benzodiazepine-affinity chromatography has been employed previously to isolate the ‘central’ benzodiazepine *To whom correspondence should be addressed

receptor from detergent-solubilized membranes (Sigel et ul., 1983; Taguchi & Kuriyama, 1984; Kirkness & Turner, 1986). Here, similar techniques resulted in the detection and puritication of a protein (P36) from the cytosolic fraction of pig cerebral cortex. An affinity column was prepared by immobilizing the benzodiazepine, 1012s. to Sepharose via a 28nm spacer arm. Following application of the cytosolic fraction from pig cerebral cortex, bound protein was eluted with the water-soluble benzodiazepine, chlorazepate. Analysis of the chlorazepate-eluate by SDS-polyacrylamide gel electrophoresis indicated a single polypeptide of M , 36000. The affinity column adsorbed P36 from the soluble fraction very effectively. Using an immunorddiometric assay, less that 5 % of the P36 loaded on to the column could be detected in the ‘run-through’ fraction. Chlorazepate was removed from the purified preparation by ultrafiltration and diafiltration followed by chromatography on a gel-filtration column. The average recovery of P36 from lOOg of cerebral cortex was 450pg and represented an enrichment, over the crude cytosolic fraction, of 3500-fold. P36 was not detected in the chlorazepate eluate of a control gel, bearing only the spacer arm. A summary of the physical properties of P36 is presented I987