[125I]-Bolton-Hunter scyliorhinin II: A novel, selective radioligand for the tachykinin NK3 receptor in rat brain

Peptides, Vol. l l, pp. 827-836. © PergamonPress plc, 1990. Printed in the U.S.A.

0196-9781/90 $3.00 + .00

[ 25I]-Bolton-Hunter Scyliorhinin II: A Novel, Selective Radioligand for the Tachykinin NK3 Receptor in Rat Brain C H R I S T I A N J. M U S S A P A N D E L I Z A B E T H

BURCHER 1

Department o f Biological Sciences, Deakin University, Victoria 3217, Australia R e c e i v e d 22 J a n u a r y 1990

MUSSAP, C. J. AND E. BURCHER. [1251]-Bolton-Hunter scyliorhinin 11: A novel, selective radioligand for the tachykinin NK3 receptor in rat brain. PEPTIDES 11(4) 827-836, 1990. --The cyclic tachykinin scyliorhinin II (SCYII) has high affinity for the [neurokinin B (NKB)-preferring] NK3 receptor. SCYII was iodinated using [~25I]-Bolton-Hunter reagent and the product BHSCYII purified using reverse phase HPLC. In rat brain membranes, binding of BHSCYII and of the relatively unselective radioligand [I25I]-Bolton-Hunter eledoisin (BHELE) was saturable, reversible and to an NK3 site. In competition studies, the rank order of potency in inhibiting binding of BHSCYII and BHELE was: SCYII - [MePheT]-NKB ~ senktide > NKB --> kassinin -> eledoisin > [Pro7]-NKB > neurokinin A > neuropeptide K --> substance P > [Sar9,Met(O2) l~]-substance P. In "cold" saturation experiments, binding of BHELE occurred to a single class of high affinity sites (KD, 18.6 _+0.91 nM). Binding of BHSCYII was of greater affinity than for BHELE and could be resolved into a high (K D, 1.33 -+0.98 nM; 27% of sites) and low affinity (KD, 9.84 + 2.75; 73% of sites) component. The total number of binding sites was similar for both radioligands (BHSCYII, 8.27 -+0.98; BHELE, 7.94 -+0.32 fmol/mg wet weight). In vitro autoradiography in slide-mounted sections of rat brain showed identical binding patterns for both radioligands (100 pM), with dense binding localized predominantly to the cortex, Ammon's horn field 1, premammillary nuclei and interpeduncular nucleus. In membranes from several peripheral tissues, specific binding of BHELE was much higher than for BHSCYII, suggesting few NK3 receptors in the periphery. In comparison with BHELE, BHSCYII exhibits lower nonspecific binding (15%), better selectivity and higher affinity for the tachykinin NK3 receptor in rat brain. Scyliorhinin II

Neurokinin B

Tachykinin receptors

Rat brain

THE tachykinin family of peptides includes substance P (SP), the mammalian decapeptides neurokinin A (NKA) and neurokinin B (NKB) and the nonmammalian eledoisin (ELE), kassinin (KASS) and physalaemin. Members of this group also include the recently discovered dogfish peptides scyliorhinin I (SCYI) and scyliorhinin II (SCYII) [(13), refer to Table 1]. The tachykinins exhibit a similar spectrum of biological activity, characterized by contraction of peripheral smooth muscle, salivation and hypotension (4,18). These actions are mediated via three distinct receptor types, now designated as NK1, NK2, and NK3 (5, 18, 12, 25). SP has the greatest potency on the NK1 receptor (potency order SP --> physalaemin > > NKA ~- KASS ELE > NKB), while the potency order at the NK2 receptor is NKA > KASS -> ELE > NKB > SP >- physalaemin and at the NK3 receptor ELE ~ KASS ~ NKB > physalaemin > NKA > SP (4,32). SP, NKA, and NKB are the putative endogenous neurotransmitters at the NK1, NK2, and NK3 receptors, respectively, although they are receptor nonselective due to their common C-terminus (4,8). In contrast to the many studies investigating the actions of SP and the extensive characterization of NK1 receptors, the distribu-

Radioligand binding

Autoradiography

tion and physiological importance of NK3 receptors have received less attention. Both functional studies and radioligand binding techniques have been used to characterize NK3 receptors in rat brain (2, 11, 12, 19, 21, 23, 38), in the guinea pig ileum (10,20), where they mediate acetylcholine release from the myenteric plexus (22), and in the rat portal vein (28). A number of useful peptide radioligands have been made using [JesI]-Bolton-Hunter (BH) reagent. The radioligand [~251]-BoltonHunter SP (BHSP), used to investigate NK1 receptors (4,8), has reasonable NK1 receptor selectivity and low nonspecific binding. Although some potent NK3 selective tachykinin analogues have recently been synthesized (17, 23, 39), no NK3 selective iodinated radioligand is currently available. NKB is relatively insoluble, and when iodinated with BH reagent it exhibits high nonspecific binding, which is quite unacceptable for autoradiography (6). In the central nervous system (CNS), investigators have employed BH-eledoisin (BHELE) as a radioligand for NK3 receptors (2, 11, 12, 16, 38). Although possessing acceptably low nonspecific binding, BHELE lacks receptor selectivity [refer to (4)] and is therefore unsuitable in tissues containing multiple tachykinin receptors. Conversely, the SP analogue " s e n k t i d e " is highly NK3

~Requests for reprints should be addressed to Elizabeth Burcher, Ph.D., School of Physiology and Pharmacology, University of New South Wales, Kensington, N.S.W., 2033, Australia.

827

828

MUSSAP AND BURCHER

TABLE 1 AMINO ACID SEQUENCES OF SEVERAL NK3 PREFERRING TACHYKININS AND ANALOGUES

Peptide

Amino Acid Sequence

NeurokininB [MePheT]-NKB Senktide Eledoisin Kassinin

Asp-Met-His-Asp-Phe-Phe-Val-Gly-Leu-Met-NH 2 Asp-Met-His-Asp-Phe-Phe-MePhe-Gly-Leu-Met-NH2 Succinyl-Asp-Phe-MePhe-Gly-Leu-Met-NH~ pGlu-Pro-Ser-Lys-Asp-Ala-Phe-Ile-Gly-Leu-Met-NHe Asp-Val-Pro-Lys-Ser-Asp-Gln-Phe-Val-Gly-Leu-Met-NH2

ScyliorhininI1

Cys-Pro-Asp-Gly-Pro-Asp-Cys-Phe-Val-Gly-Leu-Met-NH2

f

Lys-Ser-Asn-Ser-Pro-Ser

receptor selective, but the BH-labeled derivative has high nonspecific binding (21). [3H]-Senktide, used recently to label NK3 sites in brain and guinea pig ileum (15,20), exhibits appreciable levels of nonspecific binding and has low specific activity (15). It is clear that a thorough characterization and autoradiographic localization of NK3 receptors in the CNS and periphery has been hindered by the lack of a suitable, highly selective radioligand, with high specific activity and low nonspecific binding. The dogfish peptides SCYI and SCYII bind with differential selectivity to tachykinin receptors (1,7). SCYI has equal affinity for both NK1 and NK2 receptor types, while SCYII, which is the first naturally occurring cyclic tachykinin isolated, is highly NK3 selective (1,7). In this study we report, to our knowledge for the first time, the synthesis and purification of [~25I]-Bolton-Hunter scyliorhinin I1 (BHSCYII). We have investigated the binding characteristics and autoradiographic localization of binding sites for BHSCYII in rat brain, and compared them with those obtained in parallel studies using the unselective radioligand, BHELE. METHOD

Preparation and Purification of Radioligands Radioiodination of SCYII and ELE with BH reagent (specific activity 2000 Ci/mmol) was conducted as previously specified for SP (9) with some minor modifications. Specifically, peptides (50 ixg) were dissolved in 50 Ixl borate buffer (pH 9.0) and reacted with BH reagent (0.5 mCi) for 60 min at 4°C. Radioligands were purified with reverse phase HPLC, using a C-18 ix-Bondapak column (Waters) and eluted with a linear gradient of 18-50% acetonitrile (ACN, containing 0.1% trifluoroacetic acid) over 90 min, at a flow rate of 1 ml/min. Fractions corresponding to the highest levels of radioactivity were tested for binding specificity using homogenates of rat brain (refer below). Viable fractions were pooled and stored at -20°C. 13-Mercaptoethanol (1%) was added to BHELE but not to BHSCYII. Radioligands thus manufactured were usable for up to 8 weeks.

april and bestatin did not appreciably increase specific binding of either radioligand. Radioligand binding assays were conducted in parallel with 50 pM BSCYII and 80 pM BHELE. Aliquots (250 i~l) of brain homogenates (4% final concentration) were incubated at 25°C with the respective radioligand in incubation buffer. At equilibrium (BHELE, 45 rain; BHSCYII, 90 min), incubations were terminated by rapid filtration through Whatman GF/B glass fiber filters (presoaked in ice-cold 0.5% BSA). Filter-bound radioligand was washed with 50 mM Tris HC1 (pH 7.4, 4°C, containing 3 mM MnCI 2 and 0.02% BSA), dried under vacuum and quantified using a Packard Minaxi auto-gamma counter (79% efficiency). In competition studies, homogenates and radioligand were coincubated with varying concentrations of tachykinins and analogues. "Hot" saturation studies, using increasing concentrations of radioligand, were not attempted due to cost constraints. B.... and KD parameters were determined using "cold" saturation experiments (26). In such experiments, the effective radioligand concentration is increased by adding progressively larger amounts of unlabeled ligand to a fixed concentration of radioligand. In kinetic and "cold" saturation experiments, nonspecific binding for BHSCYII and BHELE was defined using 1 IxM SCYII and ELE, respectively. In competition studies, nonspecific binding for BHSCYII was defined with 1 IxM NKB and observed to be equal to that defined with 1 IxM SCYII. As a preliminary screen, homogenates from several peripheral tissues [dog bladder, 4%; rat fundus, 3%; rat duodenum (minus mucosa), 3%; guinea pig ileum (minus mucosa), 3%; guinea pig lung, 4%] were incubated in parallel with 50 pM BHSCYII and 80 pM BHELE and specific binding determined. Characterization studies were not carried out. Association and dissociation experiments were analyzed using the computer program KINETIC (26). Raw data from competition and "cold" saturation experiments were processed using EBDA (26) in conjunction with the nonlinear, iterative curve fitting program LIGAND (29).

Autoradiographic Studies Homogenate Binding Studies Sprague-Dawley rats of either sex were killed by decapitation. Their brains (minus cerebellum) were quickly removed, frozen in liquid nitrogen and stored at -80°C. Crude membrane homogenates were prepared as previously described (9) and finally suspended in incubation buffer containing 50 mM Tris HC1 (pH 7.4, 25°C), MnC12 (3 mM), bovine serum albumin (BSA; 0.02%) and the peptidase inhibitor chymostatin (4 txg/ml). In preliminary experiments, other peptidase inhibitors leupeptin, bacitracin, enal-

Frozen sections (sagittal and coronal; 10 Ixm) of rat brain were cut on a Reichert-Jung Cryocut E cryostat, thaw-mounted onto subbed glass microscope slides and stored at -20°C. When required, sections were equilibrated to room temperature, after which they were preincubated (3 × 5 min) in 170 mM Tris HC1 (pH 7.4, 25°C) containing BSA (0.02%). Sections were incubated with radioligand (100 pM) in incubation buffer containing 170 mM Tris HCI (pH 7.4, 25°C), MnC12 (3 mM), BSA (0.02%) and chymostatin (4 p~g/ml), for 120 min. Sections were then washed

SCYLIORHININ II BINDING IN RAT BRAIN

dpm (xl06)

100

A

400 ]

300

829

1

i¢

% ACN '50

y

40

/~

50

o /

30

200 -

i./"

(.9 Z 100 -

•

20

Z CO U_

10

UJ 13or)

400 -

"50

z;

300 •

"40

200 "

30

100"

20

0 0

i

i

i

i

i

i

10

20

30

40

50

60

0 0

A

10

20

, 30

, 40

, 50

I

:.%,,

0

i

0

20

40

i

60 80 TIME (min)

f 60

O/ 0

i

120

B

,oo

10

|

100

. 50

.

. 100

. . 1 50

200

250

TIME (min)

TIME (min)

FIG. 1. HPLC purification profile for BHELE (A) and BHSCYII (B). The radioactivity in each (1 min) fraction was determined. BHELE and BHSCYII (asterisked peaks) eluted at 37.5% and 33.5% ACN, respectively. Unconjugated [~251]-Bolton-Hunter reagent eluted at approximately 24% ACN.

(4 x 2 rain) in ice-cold 170 mM Tris HC1 (pH 7.4, 4°C) containing MnC12 (3 mM) and BSA (0.02%) and rinsed in ice-cold distilled water. For autoradiography, labeled sections were dried and apposed to X-ray film (Kodak) for 14 days at 4°C, before being developed as previously described (9). For biochemical studies, sections were wiped off the slide with moist Whatman GF/B glass fiber filter paper and the bound radioligand determined. Nonspecific binding for BHSCYII and BHELE was defined using 1 jxM of NKB and ELE, respectively.

Materials The peptides [MePhe7]-NKB and [Sar9,Met(O2)ll]-SP were generous gifts from Prof. D. Regoli (University of Sherbrooke, Canada), and [Pro7]-NKB was a kind donation from Dr. S. Lavielle (Paris, France). All other peptides were purchased from Peninsula Laboratories (USA) or Auspep (Australia). Standard solutions of most peptides were stored ( - 2 0 ° C ) in 0.01 M acetic acid containing 1% I~-mercaptoethanol. SCYII was stored without [3-mercaptoethanol (to maintain the disulphide bridge) and NKB was dissolved and stored ( - 20°C) in dimethylsulphoxide (DMSO).

FIG. 2. Kinetics of BHELE and BHSCYII specific binding in homogenates of rat brain. Each point in association (A) and dissociation (B) experiments represents the mean _+S.E.M. of 3-5 determinations.

BH reagent was obtained from Amersham (U.K.), while chymostatin was purchased from the Sigma Chemical Company (USA). All other reagents were of analytical grade. RESULTS

Radioligand Synthesis and Purification Chromatographic (reverse phase HPLC) purification of the radioligands produced one major peak of radioactivity (Fig. 1). This was the only fraction in each case which bound specifically to brain membranes, and represented the purified iodinated product, with the BH group presumably conjugated to the lysine side chain. BHELE and BHSCYII eluted at 37.5_+0.6% and 33.5---0.5% acetonitrile, respectively. For BHELE, this is similar to that reported by Cascieri and Liang (11). In the case of BHELE, approximately 80% of the BH label was incorporated in the peptide, while this was only about 45% for BHSCYII. Using BH reagent, we have also attempted to iodinate the NK3 receptor selective agonists [MePheV]-NKB, [MePhe7] NKB(4-10) and [Arg°,MePheV]-NKB (donated by G. Drapeau and D. Regoli, Sherbrooke University, Canada). Unfortunately, only low yields of purified product were obtained, and when binding specificity was tested in brain membranes, nonspecific binding was consistently as high as total binding. Furthermore, detergents (such as sodium dodecyl sulphate), chelating agents,

830

MUSSAP AND BURCHER

0.12"

O UA UJ rr D_

tion was about four-fold more rapid for BHELE (k_ ,, 12.8 × 10-3min - ' ; n = 5 ) than for BHSCYII (k l, 3 . 0 8 x 10-3min ~; n = 3) (Fig. 2). The kinetically derived dissociation constant (KD) for BHELE (20.7 _+4.06 pM) was greater than that determined for BHSCYII (8.89_+0.94 pM). Under these incubation conditions, binding of BHELE was 70% specific, while for BHSCYII it was 85% specific.

BHEIE CYII

006"

Z

Homogenate Binding--Saturation Studies

i

6 BOUND (fmol/mg ww)

12

FIG. 3. Representative Scatchard plots for BHELE and BHSCYII derived from 1 "cold" saturation experiment in rat brain homogenates. The Scatchard for BHSCYII is curvilinear. Bound radioligand is expressed as fmol/mg wet weight (ww) tissue. Values for BHELE: KD, 18.4 nM; B,,1,x, 8.28 fmol/mg wet weight tissue; BHSCYII: high affinity KD, 1.28, low affinity KD, 9.33 nM; B..... 5.70 fmol/mg wet weight tissue.

Binding of both radioligands was saturable (BHELE and BHSCYII, 70 nM) in homogenates of rat cerebral cortex. Hill coefficients close to unity suggest noncooperativity between sites. Scatchard plots, derived from " c o l d " saturation experiments, revealed binding of BHELE to a single population of high affinity sites (Fig. 3). In contrast, Scatchard plots of BHSCYII binding were biphasic. Simultaneous LIGAND analysis of 4 experiments revealed that BHSCYII but not BHELE binding could be resolved into two components (p = 0 ) , a high affinity (K D, 1.33 _+0.98 nM; 27% of sites) and low affinity site (K D, 9.84_+2.75 nM; 73% of sites). Specific binding of BHSCYII in rat brain was of higher affinity than that of BHELE (K D, 18.6_+0.91). The total number of binding sites labeled (Bmax) was similar for both radioligands (BHSCYII, 9.27 _+0.98; BHELE, 7.94 _+0.32 fmol/mg wet weight).

Homogenate Binding--Competition Studies antioxidants, various divalent cations and changes in buffer composition or pH failed to appreciably increase specific binding.

Homogenate Binding--Kinetic Studies Brain homogenates incubated with BHSCYII and BHELE showed saturable, reversible specific binding. Association of both BHELE (80 pM) and BHSCYII (50 pM) in homogenates of rat cerebral cortex was slow, reaching equilibrium by 45 rain and 90 min, respectively (Fig. 2). Accordingly, the association rate constant for BHELE (k+~, 6 . 8 8 x 1 0 8 M ~min ~; n = 4 ) was approximately twice that for BHSCYII (k+l, 3 . 5 4 x 10 s M lmin 1; n = 3 ) . Binding was reversible upon addition of excess (1 txM) of the corresponding unlabeled peptide. Dissocia-

Specific binding of BHSCYII and BHELE in homogenates of rat brain was characterized in parallel experiments using endogenous tachykinins and several analogues shown to have receptor selectivity. Dissociation constants (KDS) of competitors for binding were determined from the simultaneous analysis of 3-5 experiments using LIGAND (Table 2). The rank order of potency of the compounds tested in inhibiting BHSCYII binding was: SCYII -> [MePheY]-NKB --> senktide > NKB -> KASS -> ELE > [ProV]-NKB > NKA > neuropeptide K (NPK) --> SP > [Sar9,Met (O2)'I]-SP (Fig. 4). Parallel competition studies performed with BHELE showed an inhibition profile virtually identical to that for BHSCYII (Table 2). For both radioligands, the dissociation constants and potency order of competitors suggests binding to an NK3 site.

TABLE 2 BINDING PARAMETERS FOR TACHYK1NINS AND ANALOGUES AS COMPETITORS FOR BHSCYII AND BHELE BINDING IN RAT BRAIN HOMOGENATES

BHELE

BHSCYII

Slope Factor

KD (nM)

R.A.

Slope Factor

Neurokinin B Scyliorhinin II

0.91 -+ 0.06 1.03 -+ 0.02

6.75 -+ 0.84 2.83 -+ 0.21

100 241

0.94 -+ 0.04 0.86 -+ 0.02

Senktide [MePhe7]-NKB Kassinin Eledoisin [ProV]-NKB Neurokinin A Neuropeptide K Substance P [Sarg,Met(O2)n]SP

0.87 0.86 0.89 0.85 0.81 0.82 0.76 0.85 0.71

3.85 4.29 8.20 11.6 43.0 105 130 117 741

175 157 82.3 58.2 15.7 6.43 5.19 5.77 0.91

0.83 0.87 0.77 0.86 0.71 0.74 0.73 0.76 0.62

Competitor

_+ 0.11 _+ 0.03 + 0.06 ± 0.06 ± 0.05 -+ 0.07 ± 0.03 ± 0.06 + 0.01

_+ 0.22 _+ 0.89 _+ 0.51 -+ 1.2 -+ 6.9 _+ 20 ± 10 -+ 27 -+ 133

-+ 0.04 _+ 0.01 + 0.01 ± 0.02 _+ 0.01 _+ 0.05 -+ 0.01 -+ 0.08 -+ 0.01

KD (nM) 9.58 + 1.07 H: 1.37 + 0.54 L: 11.6 -+ 11.0 4.81 +_ 0.32 3.68 _+ 0.54 14.8 _+ 1.2 20.1 ± 3.1 59.9 ± 10.0 116 _+ 28.5 215 ± 20 250 -+ 34 1001 ± 151

Values are mean +- S.E.M. of 3-5 experiments. R.A., relative affinity compared with neurokinin B.

R.A. 100 699 82.5 199 260 64.6 47.7 15.9 8.26 4.46 3.83 0.96

S C Y L I O R H I N I N II B I N D I N G IN R A T B R A I N

100

xt

~'0"~. I •

lltil -

831

A

#kti

~,i.

\:

o. |~,,% • \

B

50 o

sc¥,,

•

SENK'nDE ~~1~, [MePhe'l]_NKB\-~,T

0

NKA

~,il

kl

•

NKB

"I=

~I

(.9 z

'ILx_ ~1

--\T [PRO1]'NKB

z O IJ_

O LM

0 (3. CO Z; 100 :D

i

i

k R ~ "t~

NPK

\I

I

k

I~,.

kI i

i

°,,a.Xt:l-o,,

X 3;

o,,,

-,ix,

D

\\ \g,o, \,\x

50

\%',,, \

10

!

|

i

|

I

-9

-8

-7

-6

-5

0

i

i

i

|

i

-9

-8

-7

-6

-5

LOG [COMPETITOR] (M) FIG. 4. Competition for BHELE (A, B) and BHSCYII (C, D) binding in rat brain homogenates, by various tachykinins and analogues. Each point represents the mean-+ S.E.M. of 3-6 individual experiments. SAR-SP, [Sarg,Met(O2)~I]-SP.

TABLE 3 REPRESENTATIVE dpm VALUES FOR TOTAL, NONSPECIFIC(NSB) AND PERCENT SPECIFIC (%SB) BINDING FOR BHSCYII (50 pM) AND BHELE (80 pM) IN A RANGE OF MAMMALIAN PERIPHERAL TISSUE HOMOGENATES BHSCYII

BHELE

Tissue

Total

NSB

%SB

Total

NSB

%SB

Rat brain Rat fundus Dog urinary bladder Rat duodenumt Guinea pig ileum4

10100 945 826 1259 1500

1396 765 778 890 910

86 19" 6* 29* 39*

5735 4758 2108 2118 1972

1345 793 242 720 638

76 83 88 66 68

3015

2846

6*

5189

733

86

Guinea pig lung

*Specific binding too low for accurate receptor characterization. "PMucosa removed.

Receptor Type NKI, NK3 NK2 NK2 NKI, NK2 NK1, NK2 NK3 NK 1

Reference 2, 5, 11 10 32 1,10 10,22 8

832

MUSSAP AND BURCHER

As expected, compounds shown to have high NK3 receptor selectivity, such as SCYII, [MePhe7]-NKB and senktide, were potent inhibitors of both BHELE and BHSCYII specific binding, although [ProT]-NKB (23) was less potent than other NK3selective analogues. Analysis of the SCYII competition curve against BHSCYII showed a two-site fit (Table 2). The NK2 preferring peptide NPK (1) and the NK1 receptor-selective agonist [Sar9,Met(O2) II]-SP (32) were weak competitors. Slope factors for [Sar9,Met(O2) ~I]-SP were fairly low with incomplete inhibition by this compound at 10 ~tM.

Homogenate Binding--Peripheral Tissues BHELE showed specific binding to homogenates of several peripheral tissues which contain various tachykinin receptor populations (Table 3). In rat brain, specific binding of BHSCYII was approximately twice that of BHELE binding. In contrast, negligible specific binding of BHSCYII, too low for accurate characterization, was found to rat fundus, dog bladder, rat duodenum and guinea pig lung, with marginal specific binding in guinea pig ileum. These data suggest an absence of appreciable numbers of NK3 receptors in these peripheral tissues.

Autoradiographic Studies Both BHELE and BHSCYI1 (100 pM) showed high affinity binding to slide-mounted rat brain sections. Identical patterns of binding were observed throughout the brain for both BHELE and BHSCYII (Fig. 5), with nonspecific binding very low (11%) for both radioligands (Fig. 6B, D). Very dense specific binding was associated with lamina IV of the cortex, but was much weaker over other cortical layers. The interpeduncular nucleus was very densely labeled (Fig. 5D, H), with dense binding over the premammillary nuclei (dorsal and ventral) (Fig. 5B, F) and the frontal cortex (Fig. 6). Moderate binding was observed over the olfactory tubercule, zona incerta, subiculum, substantia nigra, pars compacta and pars lateralis, peripeduncular nucleus, lateral hypothalamic area and particularly to field 1 of Ammon's horn (Fig. 5). Only faint binding was seen over the region of the central gray, subparafascicular nucleus, parvocellular part and locus coeruleus in the brainstem (not shown). Negligible binding occurred over the nucleus accumbens, fundus striati, caudate putamen, and amygdala. Notably, neither BHELE nor BHSCYII exhibited specific binding to the cerebellum or to the substantia nigra, pars reticulata. DISCUSSION In this paper we report the synthesis, purification, and binding characteristics of BHSCYII, a new selective radioligand for the NK3 receptor in rat brain. All experiments were conducted in parallel with BHELE under incubation conditions minimizing enzymatic degradation of the radioligands. Our studies show that both BHSCYII and BHELE binding to rat brain homogenates was reversible and saturable. BHSCYII exhibited exceptionally low nonspecific binding, a feature uncommon in other radioiodinated NK3 ligands, such as BHNKB (6,9) and BH-senktide [(21); C. J. Mussap, unpublished data].

Both radioligands appeared to bind to the NK3 site. Although the relative affinities (Table 2) of the competitors were comparable, the KD values of, most notably, the naturally occurring tachykinins were somewhat larger when competing for BHSCYI1, reflecting the greater affinity of this radioligand for NK3 sites. For both radioligands, the most potent competitor was SCYII, followed closely by the other NK3-selective compounds [MePheT[NKB and senktide, which were more potent than NKB itself. Generally, the rank potency orders and relative affinities of the competitors for BHELE (and BHSCYII) were in agreement with other studies in rat brain using BHELE and [3H]-NKB (2, 3, 11, 12, 23), and in guinea pig ileum and cortex using [3H]-senktide (20). In all studies, [MePheT]-NKB was more potent than NKB while eledoisin had about half and SP 1-5% of the affinity of NKB. Thus all (NK3-preferring) radioligands appear to have similar selectivity for NK3 binding sites, in contrast to the high selectivity shown by unlabeled senktide in functional studies (39). Ability to interact with binding sites is determined by affinity, whereas initiation of a response is a function of both affinity and efficacy. BHSCYII binding to NK3 receptors was of greater affinity than that of BHELE, in both saturation and kinetic experiments. However. for both radioligands, a major discrepancy exists between K D estimates determined by saturation and kinetic studies. For BHELE, such a large difference in KDs has been ascribed to problems associated with taking measurements under nonequilibrium binding conditions (2). Binding of BHELE to high and low affinity sites in rat brain has been reported (12), although in our study, as in others (2, 3, 38), BHELE bound exclusively to one population of sites. Our binding affinity for BHELE (KD, 18.6 nM) is in close agreement with estimates obtained by other investigators (2, 3, 12). Binding of BHSCYII could be resolved into two sites, of K D, 1.3 and 9.8 nM. These may represent high and low affinity states of the NK3 receptor. Alternatively, our high affinity site may represent a functional component of the receptor while the lower affinity site may be an interactive binding site. Binding of BHSCYII to CNS NKI receptors is most unlikely, since the autoradiographic distribution of binding sites for BHSCYII (and BHELE, as shown here) in rat CNS is quite different from the pattern of binding seen with BHSP (6, 16, 33). Furthermore, NKl-selective competitors were extremely weak in inhibiting binding of BHSCYII when compared to the potency of NK3selective compounds. The high affinity K D of BHSCYII in this study is in good agreement with values found for BHELE using the cloned NK3 receptor [0.64 nM (35)], in rat CNS for BH-senktide [0.9 nM (21)], [3H]-senktide [2.8 nM (15)], BHELE [0.63 nM (19); 0.9 nM (12)] and [3H]-NKB [KD, 4 nM (3)1, and for [3Hl-senktide in guinea pig ileum [2.2 nM (20)1. SCYII, senktide, NKB and its analogues share certain structural features, such as possession of aspartic acid at position 4 (relative to NKB; refer to Table 1), which are important in both binding affinity and interaction at the NK3 receptor (17,31). Tachykinins and analogues lacking this aspartic acid residue {for example, SP and [Sarg,Met(O2) It ]-SP}, or containing lysine, were weaker competitors for binding to NK3 sites. Possession of a disulphide bridge appears to significantly enhance NK3 selectivity and affinity. The cyclic analogue [Cys2"5] -

FACING PAGE FIG. 5. Photomicrographs taken directly from X-ray film images of total binding of BHSCYII (A, B, C, D) and BHELE (E, F, G, H) to rat brain sections. (A, E): IA (interaural line), 13.7 mm; (B, F): IA, 7.0 ram; (C, G): IA, 3.2 mm; (D, H): IA, 2.2 mm. AO=posterior accessory olfactory bulb, CA 1 = Ammon's horn (field 1), IP = interpeduncular nucleus, CG = central gray, DG = dentate gyrus, Fr = frontal cortex, PM = premammillary nuclei, SNC = substantia nigra, pars compacta, SNR = substantia nigra, pars reticulata, Str = striate cortex. Bar represents 3 mm.

SCYLIORHININ II BINDING IN RAT BRAIN

833

F

B

/, / DG

Str

PM .~,

C

SNR tq

,.

CA1

834

MUSSAP AND BURCHER

A '-

B Fr

/

Str

/

SNC-

C

LH I

FIG. 6. Photomicrographs taken directly from X-ray film images of sagittal rat brain sections labeled with BHSCYII (A, B) and BHELE (C, D). Sections shown in B and D were coincubated with I IxM NKB, and represent nonspecific binding. Ce = cerebellum, LH = lateral hypothalamic area, with other abbreviations as described for Fig. 5. Bar represents 3 ram.

NKB and SCYII have high affinity for the NK3 receptor (1,23). Interestingly, scyliorhinin II(7-18) has even greater NK3 receptor affinity and selectivity than the full length molecule (7). Disulphide bridging may cause conformational changes at the Nterminus of the molecule, thereby permitting the carboxy side chains of the aspartic acids (positions 7 and 13 for SCYII) greater access to complementary cationic binding sites. The NK3 receptor has only recently been cloned, and shows considerable homology with the NK1 and NK2 receptors (35). One can speculate that the NK3 receptor may have two distinct binding sites for tachykinins, one [suggested to be the conserved transmembrane region VII (35)] recognizing the common C-terminus, with the other having high affinity for the anionic aspartic acid residues present in NK3-preferring compounds. An understanding of the position and amino acid sequence of binding sites within the receptor will enable the development of more selective agonists and antagonists. Autoradiographic localization of binding sites for both BHELE and BHSCYII in coronal and sagittal sections from different layers of rat brain appeared of identical distribution, suggesting binding to the same class of sites. This conclusion is further supported by comparable Bmax values for BHELE and BHSCYII. The distribution of binding sites for both radioligands was in good agreement with the autoradiographic patterns of BHELE and [3H]-NKB binding obtained previously in rat CNS (6, 16, 33, 34). Dense BHSCYII binding was observed in layer IV of various regions of the cerebral cortex. The presence of moderate binding in the substantia nigra, pars compacta, and the absence of sites in the pars reticulata are in good agreement with data of Saffroy and

coworkers using [3H]-NKB and BHELE (34) and with the study of Dam and coworkers using [3H]-senktide (15). Dense binding to the interpeduncular and premammillary nuclei was particularly noticeable and in agreement with previous studies (6,34). In our study, binding in the paraventricular nucleus was specific and of low density, in contrast to the high, nonspecific binding of [3H]-senktide in this region (15). We also found appreciable BHSCYII and BHELE binding to the subiculum, in contrast to the study with [3H]-senktide (15). This last discrepancy is not easy to explain, although Dam and coworkers suggest that BHELE may be binding weakly to NK1 sites. BHELE is considered unselective in the periphery (see below), but in our study, no specific binding occurred in areas (e.g., striatum, dorsal CA3 field) shown previously to contain NK1 receptors (6, 16, 34), suggesting that both iodinated radioligands show good selectivity for NK3 receptors in the rat brain. NK3 receptors may mediate several functions in the CNS. The presence of NK3 sites in the supraoptic nucleus (6,34) may be correlated with levels of NKB-like immunoreactivity in that region, and the possible role of NKB in central pressor mechanisms (30). NK3 agonists are potent in inducing antinatriorexic effects (27), while the initiation of chewing, yawning and sexual arousal by senktide may be due to NK3 receptor-mediated stimulation of central cholinergic mechanisms (36). High levels of NKB-like immunoreactivity in the hypothalamus (37) may be of relevance to these actions. However, precise "matching" of NK3 binding sites to NKB-immunoreactive terminals is not possible at this time, due to an absence of highly specific NKB antibodies and comprehensive immunohistochemical studies in the CNS.

SCYLIORHININ II BINDING IN RAT BRAIN

835

In the periphery, there are only limited reports of NK3 receptors (10, 22, 28) and generally negative reports of NKB-like immunoreactivity. In our preliminary studies using homogenates from peripheral tissues, BHELE bound with high affinity to membranes of tissues containing NK1 and/or NK2 receptors (Table 3), whereas BHSCYII generally showed minimal specific binding, except in guinea pig ileum. In peripheral tissues, BHELE appears to be an unselective radioligand, binding to definitively characterized NK1 and NK2 sites (9, 10, 24). The adult rat brain lacks appreciable numbers of NK2 receptors (6,14), but BHELE would be expected to bind to both the NK1 and NK3 sites present. However, BHELE appears to have negligible affinity for NK1 sites in CNS but does recognize peripheral NK1 and NK2 receptors [(9, 10, 24); refer to Table 3]. Several conclusions derive from these data. Firstly, BHSCYII has much lower affinity than BHELE for NK1 and particularly for NK2 peripheral tachykinin receptors, corresponding with the differential receptor affinity of the unlabeled peptide (7). Secondly, few populations of NK3 receptors appear to exist in the peripheral tissues investigated, except for guinea pig ileum where NK3 receptors mediate acetylcholine release from myenteric plexus neurons (22). NK3 binding

sites have been demonstrated in myenteric plexus/longitudinal muscle homogenates using BHELE (10) and more recently, Guard and coworkers (20) have characterized high affinity [3H]-senktide binding to NK3 sites in this tissue. Finally, NK1 binding sites in brain may be different from NK1 sites in peripheral tissues, since BHELE shows negligible affinity for brain and good affinity for peripheral NK1 binding sites (9, 10, 23). In this study we have shown that BHSCYII is a novel, selective radioligand for the tachykinin NK3 receptor in rat brain. BHSCYII appears to be superior to BHELE, in at least three important respects. It has a much greater affinity for brain NK3 receptors, it has greater selectivity, and, very importantly, it exhibits less nonspecific binding. Our results suggest that BHSCYII will be a useful and selective tool for the characterization and autoradiographic localization of tachykinin NK3 receptors. ACKNOWLEDGEMENTS This study was supported by a grant to E.B. from the National Health and Medical Research Council of Australia. We thank Russell Sharp for photographic printing, George Paxinos for advice on neuroanatomy and Dominic Geraghty for helpful comments.

REFERENCES 1. Beaujouan, J.-C.; Saffroy, M.; Petitet, F.; Torrens, Y.; Glowinski, J. Neuropeptide K, scyliorhinin I and II: New tools in the tachykinin receptor field. Eur. J. Pharmacol. 151:353-354; 1988. 2. Beaujouan, J.-C.; Torrens, Y.; Viger, A.; Glowinski, J. A new type of tachykinin binding site in the rat brain characterized by specific binding of a eledoisin labeled derivative. Mol. Pharmacol. 26: 248-254; 1984. 3. Bergstrom, L.; Torrens, Y.; Saffroy, M.; Beaujouan, J.-C.; Lavielle, S.; Chassaing, G.; Morgat, J.-L.; Glowinski, J.; Marquet, A. 3Hneurokinin B and ~251-Bolton-Hunter eledoisin label identical tachykinin binding sites in the rat brain. J. Neurochem, 48:125-133; 1987. 4. Buck, S. H.; Burcher, E. The tachykinins: A family of peptides with a brood of "receptors." Trends Pharmacol. Sci. 7:65~8; 1986. 5. Buck, S. H.; Butcher, E.; Shults, C. W.; Lovenberg, W.; O'Donohue, T. L. Novel pharmacology of substance K-binding sites: A third type of tachykinin receptor. Science 226:987-989; 1984. 6. Buck, S. H.; Helke, C. J.; Butcher, E.; Shults, C. W.; O'Donohue, T. L. Pharmacologic characterization and autoradiographic distribution of binding sites for iodinated tachykinins in rat central nervous system. Peptides 7:1109-1120; 1986. 7. Buck, S. H.; Krstenansky, J. L. The dogfish peptides scyliorhinin I and scyliorhinin II bind with differential selectivity to mammalian tachykinin receptors. Eur. J. Pharmacol. 144:109-111; 1987. 8. Burcher, E. The study of tachykinin receptors. Clin. Exp. Pharmacol. Physiol. 16:539-543; 1989. 9. Burcher, E.; Buck, S. H. Multiple tachykinin binding sites in hamster, rat and guinea-pig urinary bladder. Eur. J. Pharmacol. 128:165-177; 1986. 10. Burcher, E.; Buck, S. H.; Lovenberg, W.; O'Donohue, T. L. Characterization and autoradiographic localization of multiple tachykinin binding sites in gastrointestinal tract and bladder. J. Pharmacol. Exp. Ther. 236:819-831; 1986. 11. Cascieri, M. A.; Liang, T. Binding of [~25I]Bolton-Hunter conjugated eledoisin to rat brain cortex membranes--evidence for two classes of tachykinin receptors in the mammalian central nervous system. Life Sci. 35:179-184; 1984. 12. Cascieri, M. A.; Chicchi, G. G.; Liang, T. Demonstration of two distinct tachykinin receptors in rat brain cortex. J. Biol. Chem. 260:1501-1507; 1985. 13. Conlon, J. M.; Deacon, C. F.; O'Toole, L.; Thim, L. Scyliorhinin I and II: two novel tachykinins from dogfish gut. FEBS Lett. 200: 111-116; 1986. 14. Dam, T. V.; Escher, E.; Quirion, R. Evidence for the existence of three classes of neurokinin receptors in brain. Differential ontogeny of neurokinin-l, neurokinin-2 and neurokinin-3 binding sites in rat cerebral cortex. Brain Res. 453:372-376; 1987.

15. Dam, T.-V,; Escher, E.; Quirion, R. Visualization of neurokinin-3 receptor sites in rat brain using the highly selective ligand [3H]senktide. Brain Res. 506:175-179; 1990. 16. Danks, J. A.; Rothman, R. B.; Cascieri, M. A.; Chicchi, G. G.; Liang, T.; Herkenham, M. A comparative autoradiographic study of the distributions of substance P and eledoisin binding sites in rat brain. Brain Res. 385:273-281; 1986. 17. Drapeau, G.; D'Orleans-Juste, P.; Dion, S.; Rhaleb, N. E.; Regoli, D. Specific agonists for neurokinin B receptors. Eur. J. Pharmacol. 136:401-403; 1987. 18. Erspamer, V. The tachykinin peptide family. Trends Neurosci. 4:267-269; 1981. 19. Foster, A. C.; Tridgett, R, Comparison of the binding of radiolabelled neurokinin A and eledoisin in rat cortex synaptic membranes. Br. J. Pharmacol. 94:602-608; 1988. 20. Guard, S.; Watson, S. P.; Maggio, J. E.; Phon Too, H.; Watling, K. J. Pharmacological analysis of [3H]-senktide binding to NK3 tachykinin receptors in guinea-pig ileum longitudinal muscle-myenteric plexus and cerebral cortex membranes. Br. J. Pharmacol., in press. 21. Laufer, R.; Gilon, C.; Chorev, M.; Selinger, Z. Characterization of a neurokinin B receptor site in rat brain using a highly selective radioligand. J. Biol. Chem. 261:10257-10263; 1986. 22. Laufer, R.; Wormser, U.; Friedman, Z. Y.; Gilon, C.; Chorev, M.; Selinger, Z. Neurokinin B is a preferred agonist for a neuronal substance P receptor and its action is antagonized by enkephalin. Proc. Natl. Acad. Sci. USA 82:7444-7448; 1985. 23. Lavielle, S.; Chassaing, G.; Ploux, O.; Loeuillet, D.; Besseyre, J.; Julien, S.; Marquet, A.; Convert, O.; Beaujouan, J.-C.; Torrens, Y.; Bergstrom, L.; Saffroy, M.; Glowinski, J. Analysis of tachykinin binding site interactions using constrained analogues of tachykinins. Biochem. Pharmacol. 37:41-49; 1988. 24. Lee, C. M.; Campbell, N. J.; Williams, B. J.; Iversen, L. L. Multiple tachykinin binding sites in peripheral tissues and brain. Eur. J. Pharmacol. 130:209-217; 1986. 25. Lee, C. M.; Iversen, L. L.; Hanley, M. R.; Sandberg, B. E. B. The possible existence of multiple receptors for substance P. Naunyn Schmiedebergs Arch. Pharmacol. 318:281-287; 1982. 26. McPherson, G. A. A practical computer-based approach to the analysis of radioligand binding systems. Comp. Prog. Biomed. 17:107-114; 1983. 27. Massi, M.; Polidori, C.; Perfumi, M.; Gentili, L.; De Caro, G.; Maggi, C. A. Tachykinins and salt appetite: A possible role for central NK-3 receptors. Regul. Pept. 22:123; 1988. 28. Mastrangelo, D.; Mathison, R.; Huggel, H. G.; Dion, S.; D'OrleansJuste, P.; Rhaleb, N. E.; Drapeau, G.; Rovero, P.; Regoli, D. The rat isolated portal vein: A preparation sensitive to neurokinins, particu-

836

larly neurokinin B. Eur. J. Pharmacol. 134:321-326; 1986. 29. Munson, P. J.; Rodbard, D. LIGAND: A versatile computerized approach for characterization of ligand-binding systems. Anal. Biochem. 107:220-239; 1980. 30. Nagashima, A.; Takano, Y.; Tateishi, K.; Matsuoka, Y.; Hamaoka, T.; Kamiya, H. Central pressor actions of neurokinin B: Increases in neurokinin B contents in discrete nuclei in spontaneously hypertensive rats. Brain Res. 499:198-203; 1989. 31. Osakada, F.; Kubo, K.; Goto, K.; Kanazawa, I.; Munekata, E. The contractile activities of neurokinin A, B and related peptides on smooth muscles. Eur. J. Pharmacol. 120:201-208; 1986. 32. Regoli, D.; Drapeau, G.; Dion, S.; D'Orleans-Juste, P. Pharmacological receptors for substance P and neurokinins. Life Sci. 40: 109-117; 1987. 33. Rothman, R. B.; Danks, J. A.; Herkenham, M.; Cascieri, M. A.; Chicchi, G. C.; Liang, T.; Pert, C. B. Autoradiographic localization of a novel peptide binding site in rat brain using the substance P analog, eledoisin. Neuropeptides 4:343-349; 1984. 34. Saffroy, M.; Beaujouan, J.-C.; Torrens, Y.; Besseyre, J.; Bergstrom,

MUSSAP AND BURCHER

35.

36.

37.

38.

39.

L.; Glowinski, J. Localization of tachykinin binding sites (NK~, NK 2, NK 3 ligands) in the rat brain. Peptides 9:227-241; 1988. Shigemoto, R.; Yokota, Y.; Tsuchida, K.; Nakanishi, S. Cloning and expression of a rat neuromedin K receptor cDNA. J. Biol. Chem. 265:623~528; 1990. Stoessl, A. J.; Dourish, C. T.; Iversen, S. D. Behavioural evidence for CNS tachykinin-cholinergic interactions. Regul. Pept. 22:178; 1988. Tateishi, K.; Matsuoka, Y.; Hamaoka, T. Establishment of highly specific radioimmunoassays for neurokinin A and neurokinin B and determination of tissue distribution of these peptides in rat central nervous system. Regul. Pept. 24:245-257; 1989. Torrens, Y.; Lavielle, S.; Chassaing, G.; Marquet, A.; Glowinski, J.; Beaujouan, J.-C. Neuromedin K, a tool to further distinguish two central tachykinin binding sites. Eur. J. PharmacoI. 102:381-382; 1984. Wormser, U.; Laufer, R.; Hart, Y.; Chorev, M.; Gilon, C.; Selinger, Z. Highly selective agonists for substance P receptor subtypes. EMBO J. 5:2805-2808; 1986.