Analog specificity of the thyrotropin-releasing hormone receptor in the central nervous system: Possible clinical implications

Life Sciences, Vol. 36, pp. 601-611 Printed in the U.S.A.

Pergamon Press

ANALOG SPECIFICITYOF THE THYROTROPIN-RELEASING HORMONE RECEPTOR IN THE CENTRAL NERVOUSSYSTEM: POSSIBLECLINICALIMPLICATIONS. E .F. Hawkins and W.I~. Engel Neuromuscular Center, University of Southern California School of Medicine 637 South Lueas Avenue, Los Angeles, CA 90017-1912 (Received in final form November 27, 1984)

SUMMARY TRH has rapid-onset (30 see), slow-offset (1-12 days) clinical benefit in patients with amyotrophie lateral sclerosis and other motor neuron disorders. This benefit is probauly receptor-mediated and may nave at least 2 components. To obtain a better understanding of the various responses to TRH of the spinal lower motor neurons (LMNs) in patients, and possibly to help guide selection of additional therapeutic agents, we utilized rat CNS (spinal-cord and brain membranes) to analyze the ability of certain molecules to inhibit specific binding of [3H]methyl TRH ([3H]MeTRH) to the TRH receptor. We found: a) lack of nigh-affinity binding of tt~e TRH-analog DN1417 by spinal-cord and brain TRH receptor, despite its known strong TRHlike action physiologically on LMNs; b) lack of ~ign-atfinity ~inding of the TRH-product cyclo(His-Pro) by spinal-cord and brain TRH receptor despite its havlngsome strong TRH-like pnysiologic actions on the CNS; and c) lack of any identifiable high-affinity receptor for cyclo(His-Pro) in spinal cord and brain. From these data we hypothesize that the acute transmltter-liKe action of DN-1417, TRH, and possibly other TRH-analogs and products on LMNs is via a non-TRH receptor, such as an amine or amino acid neurotransmitter receptor, e.g. a 5-hydroxytryptamine receptor. We further postulate that tne CNS TRH-receptor may modulate a tropnic-like influence of TRH on LMNs. TRH has rapid-onset, slow-offset clinical benefit in patients with amyotropnic lateral sclerosis and other motor neuron disorders tl-3). This benefit, wnicnis probably receptor-mediated, may have at least 2 components, possibly attributable to one or more receptor types and/or cellular sites. Acutely (onset 30 sec) tnere is increased lower motor neuron (LMN) excitation associated with increased strength, and subacutely (onset 1-3 hrs) there is, in addition and seemingly paradoxically, decreased spasticity apparently associated with lessened LMN excitation (1-3). Offset of both beneficial clinical effects can be as long as 1-12 days (3). With excess dosage, a clinically evident autorefractory state can occur, characterized by rapid loss of strength improvement and some increased weakness (4). After i.v. TRH it lasts a median of 68 minutes, or after chronic excess subcutaneous dosage as long as 24-48 hr (WKE, unpublished observation). The cellular mechanisms of these rapid transmitter-liKe and more cnronic trophic-like responses are not known. Animal studies with TRH have shown acute transmitter-liKe excitation of LMNs (510), possibly acting by facilitating glutamate (10,11) and/or aspartate excitation, an acute autorefractory state (5,7,10), and a longer-term trophic-like action on tissueculturea LMNs (12). Two analog~ of TRH (e.g. DN-1417; R-~ 77368) nave been demonstrated to mimic its LMN excitation (8,13), and some TRH-products (e.g. histidineproline diKetopiperazine (cyclo(His-Pro)) {14,15) and deamido-TRH (15)) mimic certain central nervous system (CNS) actions of TRH. A TRH receptor exists in the CNS (16-21) but its role in relation to LMN function regarding TRH, TRH-analogs, TRH-products, and other possibly relevant molecules is not known. 0024-3205/85 $3.00 + .00 Copyright

(c) 1985 Pergamon Press Ltd.

602

Analog Specificity of CNS TRH Receptors

Vol. 36, No. 6, 1985

In order to obtain a better understanding of these various responses at, or impinging upon, the spinal LMNs in patients, and possibly to help guide selection of additional therapeutic agents, we h a v e utilized rat CNS (spinal-cord and brain membranes) to analyze the ability of certain potentially relevant molecules to inhibit the specific binding of [3H]methyl TRH ([3H]MeTRH) to the TRH receptor. We have also looked for a high-affinity receptor for eyclo (His-Pro) since it is possible that this metabolite, which we found does not bind to the CNS TRH receptor, might be independently responsiDle for some of the effects observed in ALS patients following TRH administration. MATERIALS AND METHODS Materials. Tritiated [3-methyl-histidyl2]-tnyrotropin-releasing hormone ([3H]MeTRH; 7"~ Ci~mol) and [3H]cyclo-1 histidyl-I proline ([3H]cyelo(His-Pro); 56.7 Ci/mmol) were purchased from New England Nuclear, Boston, MA. Unlabeled MeTRH, TRH, deamidoTRH(TRH-OH), [Glul]-TRH, pGlu-His, cyclo (His-Pro), somatostatin, substance P, CCKoctapeptide and leuteinizing hormone-releasing hormone (LH-RH) were obtained from Peninsula Laboratories, Belmont, CA. The following TRH analogs were generous gifts; DN-1417 (TaKeda Chemical Industries, Japan), MK-7?I (Merck Sharp and Dohme, USA), and RX77368 (Reckitt and Colman, UK). All other compounds were purchased from Sigma Chemical Co., St. Louis, MO. Tissue Collection. Brains and spinal cords were harvested from male Sprague-Dawley rats (280-300 g m ) . The animals were sacrificed by decapltation, the brains rapidly removed, rinsed in ice-cold saline, blotted dry, and placed on a glass plate cooled by ice. The combined pyriform cortex/amygdala (PC/A) regions were aissected out (22) and stored frozen in liquid nitrogen. The entire spinal cord minusnerve roots was dissected out, rinsed, blotted dry, and stored frozen in liquid nitrogen. Preparation of crude membrane suspensions. Unlessotherwise indicated, all procedures were performed in a 4°C cold room. Frozen PC/A or spinal cord was pulverized at liquid nitrogen temperature using a specially constructed stainless steel pestle and mortar. The resulting powder was mixed with 4 volumes of 20 mM potassium phosphate buffer, pH 7.4 at 0°C (hereafter referred to as "buffer") and homogenized in ice using a Polytron PT 10ST tissue homogenizer (Brinkmann Instruments Inc., Westbury, NY). The nomogenate was centrifuged at 21,000 rpm (30,000 x g) in a Beckman Type 65 rotor for 20 rain at 2°C. The supernatant was discarded, and the pellet was surface-washed with 2 ml Duffer and resuspended in buffer with a glass Dounee homogenizer and an "A" (loosefitting) pestle. The tissue:buffer ratio for resuspension was 80 mg original tissue wt per ml buffer for PC/A and 150 mg per ml for spinal cord. Equilibrium i)indingexperiments. In "on-rate" experiments, conditions were determined for the equilibrium saturable binding of [3H]MeTRH to PC/A membranes. Membrane suspensions were incubated at 0°C with a final [3H]MeTRH concentration of 6.5 nM in the absence and presence of 6.5 ~M of the unlabeled hormone. These incubates were used to determine total and non-specific binaing respectively from which specific receptor-bound hormone was determined by difference. Periodically, duplicate 200 IJ1 ahquots of the incubates were withdrawn ana free hormone removed by vacuum filtration on Whatman GF/B filters. The filters were washed 3 times with 2 ml buffer and transferred to plastic scintillation vials for counting in I0 ml Scintiverse II (Fisher Scientific). They were left to extract overnight prior to dpm determination in a Packard too(tel 2660 liquid scintillationcounter. The rate of displacement of receptor bound [3H]MeTRH by unlabeled hormone was determined in "off-rate" experiments that included an additional tube containing membrane suspension and [3H]MeTRH to which 6.5 IJM of unlabeled MeTRH was added once equilibrium bindingwas established (designated as time 0). This mixture, together with incubates for total and non-specific binding determination, were sampled at various time intervals thereafter and free hormone removed by vacuum filtration and washing as described above. Bounddpm were determined by liquid scintillationcounting.

Vol. 36, No. 6, 1985

Analog Specificity of CNS TRH Receptors

603

Titration analyses. To determine the affinity and binding-site concentration of the CNS TRH receptor, duplicate 200 IJ1 aliquots of PC/A and spinal cord membrane suspensions were incubated for 6 hr at 0°C with [3H]MeTRH at concentrations ranging from 0.05 to 13 nM in the absence (total binding) or presence (non-specific binding) of a 1000-fold molar excess of unlabeled MeTRH. The final incubate volume was adjusted to 400 I~1 with buffer. To terminate the incubation and remove free hormone, 2 ml ice cold buffer were added to each incubation tube and the tube contents vacuum-filtered through GF/B filters. Filter washing and scintillation counting were as described above. The data were plotted according to the method of Scatchard (23) and lines were fitted to the data points by least-squares regression analysis. Competition experiments. Ligandspecificity of the TRH receptor in PC/A and spinal cord was determined by incubating duplicate 200 ~I aliquots of membrane suspensions for 6 hr at 0°C with [3H]MeTRH alone at 6.5 nM (total binding) or in the additional presence of increasing molar excesses of TRH-analogs and other eompounas. The volume of each incubate was adjusted to 400 pl with buffer. Specific (receptor) binding was measured in each e x p e r i m e n t by including t r i p l i c a t e incubates containing [3H]MeTRH and i3 ~M unlabeled MeTRH (determination of non-specific binding). Incubations were t e r m i n a t e d by vacuum f i l t r a t i o n and r a d i o a c t i v i t y was d e t e r m i n e d as described above for t i t r a t i o n e x p e r i m e n t s . The c o n c e n t r a t i o n of c o m p e t i t o r required for 50% inhibition of specific [3H]MeTRH binding (IC50) was determined from plots of receptor-bound [3H]TRH as a function of log molar c o m p e t i t o r c o n c e n t r a t i o n . These IC50 values were used to calculate ki values of competitors where Kin=IC50/I + C/Kd (Ki and Kd = equilibrium dissociation constant of competitor and of [~H]MeTRH respectively for PC/A or spinal cord TRH receptor; C = concentration of [3H]MeTRH (6.5 nM)). Binding studies w i t h [3H]cyelo(His-Pro). We verified the purity and identity of [3H]cyclo(His-Pro) by thin-layer chromatography on cellulose polyethyleneimineimpregnated plates using the solvent system n-butanol:ethyl acetate:acetic acid:water (1:1:1:1). "On-rate" and Scatcnard plot titration experiments were performed with crude membrane suspensions, prepared as described above f r o m whole spinal cords, nypothalamus and extrahypothalamic brain (minus amygdala, pyriform cortex and cerebellum). For "on-rate" experiments, membrane suspensionswere incubated with 17.5 nM [3H]cyclo(His-Pro) in the absence and presence of a 1500-fold molar excess of unlabeled cyclo(His-Pro) for periods ranging from 10 to 120 minutes at 37°C or 15 minutes to 24 hours at 0°(3. Periodically, duplicate aliquots were withdrawn from each incubation mixture and specific binding was determined as described above. Scatchard plots were performed on membrane preparations with [3HS eyclo(HisPro) concentrations ranging from 0.7 to 175 nM in the absence and presence of 13 ]~M unlabeled cyclo(His-Pro). Incubations, in auplicate for each ligand concentration, were continued for 6 hours at 0°C. They were terminated by vacuum filtration and specific binding was determined Dy scintillationcounting. Additional titration experiments were performed with high-speed (105,000 x g) supernatants (cytosols) from these three tissues. Two hundred microliter aliquots of cytosol were incubated in duplicate with [3H] cyclo(His-Pro) at identical concentrations to those used in the membrane experiments, for 6 hours at 0°C. All incubates were adjusted to a final volume of 400 ~l with buffer. Free and bound peptide were separated by chromatography of the cytosols on calibrated Sepnaaex G-25 minicolumns. Fractions of the void volumes were collected and counted and the data were analyzed for specific [3HScyclo(His-Pro) binding according to the method of Cnamness and McGuire (24). Protein determinations. Protein assays were performed by the method of Lowry e t al. (2 5) using bovine serum albumin as the standar(l.

604

Analog Specificity of CNS TRH Receptors

Vol. 36, No. 6, 1985

RESULTS Our results were obtained with crude, unfraetionated, "membrane" suspensions. In two pilot experiments utilizing isopycnic centrifugation techniques to separate subceUular components of rat brain PC/A, we found that only 6% of total cellular TRH receptor activity was recoverable in cytosol. The high-affinity TRH binding sites were particularly enriched in the partially-purifiedmembrane fraction (approximately 60% of total), followed by nuclei (19%) and microsomes (14%), with negligible activity in the mitochondrial/lysosom al pellet. Equilibrium binding experiments. Specific binding of [3H]MeTRH to PC/A membranes reached equilibrium after 2 hours at 0°C and remained unchanged for up to 8 hours of incubation (Fig. 1A). Non-specific binding was 16-19% of the total binding measured. At the next sampling interval (22 hours), specific binding had deelined by 20% (data not shown). In "off-rate" experiments, displacement of labeled [3H]MeTRH by the unlabeled hormone was relatively rapid, with a half-time of approximately three hours (Fig. 1B). These data suggest that the receptor sites that we are measuring are probably unoccupied by endogenous TRH, since there was no significant increase in specific [3H]MeTRH binding beyond two hours of incubation (Fig. 1A). As standard time and temperature conditions for labeling of receptor sites in subsequent experiments, we have used 6 hours at 0°C.

16 t I

A.

o x v

O z r~10 Z m T 8 rr

1 i

10

TIME (hou,s) 3 4 ~ i

2 L

5 I

6 =

1 I

090807

kI

0605O/O o 04-

2--~ 4 w Q. c~

03-

2

I

I

I

0

1

2

I

I

I

3 4 5 TIME (hours)

I

I

I

6

7

8

FIG. 1. A: Uptake of [3H]MeTRH by rat PC/A membrane preparations. Membrane suspensions were incubated with [3H]MeTRH at 6.5 nM in the absence and presence of 6.5 ]JM of unlabeled MeTRH. Periodically, duplicate 200 pl aliquots of each incubate were withdrawn and free hormone removed by vacuum filtration on Whatman GF/B filters. B: Displacement of receptorbound [3H]MeTRH by unlabeled hormone. After three hours of incubation (designated as time 0), unlabeled MeTRH was added at a final concentration

Vol. 36, No. 6, 1985

Analog Specificity of CNS TRH Receptors

605

of 6.5 ~M to a third tube containing membrane suspension and [3H]MeTRH. This mixture together with incubates for total and non-specific bindingwere then sampled at various time intervals and free hormone removed by vacuum filtration. The values shown are the means ancl standard deviations of two separate experiments. Titration experiments. Scatchard plots of titration data have revealed an apparently single class of similar, high-affinity, saturable [3H]MeTRH binding sites in PC/A and spinal cord membranes. A representative spinal cord Seatchard plot is illustrated in Figure 2. Kd values were similar in both tissues (Kd in PC/A = 2.0 + 1.0 nM, Kd in spinal cord = 4.7 + 1.6 nM; means _+ SD of 8 and 4 separate determinations respectively) and are in agreement with values reported by others for [3H]MeTRH binding in the mammalian CNS (17,19,21). The concentration of specific binding sites was approximately one order of magnitude lower in spinal cord than in PC/A (7.6 + 1~3 vs. 104 + 8.3 femtomoles/mg protein for 4 and 8 determinations respectively). Additionally the concentration of low-affinity, non-specific sites was substantially higher in spinal cord than in PC/A (45% vs. 1896 of total binding). 0.03

•

Kd = 3.96x10-gM Bmax

= 7.9 F Moles/mg

protein

r = -0.957

0.02

B/F

0.01

0.00 0

I

I

I

|

2

4

6

8

SPECIFIC [3HI Me-TRH BINDING (femtomoleslmg protein) FIG. 2 Representative Seatchard plot of [3H]MeTRH binding to rat spinal-cord membranes. Membranes were incubated in duplicate with [3H]MeTRH concentrations ranging from 0.05 to 13 nM in the absence (total bincling) and presence (non-specific binding) of a 1000-fold molar excess of unlabeled MeTRH for 6 hours at 0 ° C . Incubations were terminated by vacuum filtration. Competition experiments. To date, 33 compoundshave been tested for their abilities to inhibit [aH]MeTRH binding to the TRH receptor in PC/A membranes (Table I). Over the concentration ranges tested, only six of these, in addition to TRH, exhibited significant competitive activity and all six were structurally related to TRH (Table 2; Fig. 3).

606

Analog Specificity of CNS TRH Receptors

Vol. 36, No. 6, 1985

TABLE 1 Compounds Tested in Brain TRH Receptor Specificity Analyses

TRH/analogs/m etabolites

5-HT Antagonistsb

Others a

MeTRHa TRHa Cyclo (His-Pro)b TRH-OHc [Glul]-TRHb p-glu-hisa MK-771b DN-1417b RX77368 b

Cyproheptadine Methysergide Spiperone Ketanserin

l-pyroglutamic acid d, l and dl-glutamic acids 1-glutamine Glycine GABA 5-HT Substance P LH-RH Somatostatin CCK-octapeptide Quinineb

Lectins d Con A Wheatgerm Soybean

Dopamine agonistse Apomorptline Bromocryptine 3-Hydroxytyramine L-dopa

a = tested up to 13 pM b = tested up to 65 pM c = tested up to 325 IJM

d = tested up to 200 IJg e = tested up to 130 IJM

Membranes were incubated with [3H]MeTRH at 6.5 nM and increasing molar excesses of competitors for 6 nours at 0°C. Unbound hormones were removed by vacuum filtration on Whatman GF/B filters. Over the concentration ranges tested, only the following compounds, in adaition to TRH, had significant competitive activity: MeTRH, MK-771, RX77368, [Glul]-TRH, DN-1417, TRH-OH.

TABLE 2. [3H]MeTRH binding to the CNS TRH receptor: Effect of Competitors.

(nM) Compound

Brain

Spinal Cord

MeTRH TRH MK-771 RX 77368 [Glul]-TRH DN-1417 TRH-OH

2 ~3 28.2 212 282 2820 4230 28200

4.4 54 122 259 7 570 6924 267 50

IC 50 Ki

-

-

-

C

1+~-~

where Ki and Kd = equilibrium dissociation constant of competitor and [3H]MeTRH respectively. C = concentration of [3H]MeTRH (6.5 nM).

Vol. 36, No. 6, 1985

Analog Specificity of CNS TRH Receptors

-- -o---o * * o--o

lOO z

_z m "l-

n-

o

Me-TRH TRH DN1417 RX77368

607

~ MK-?71 .--* TRH-OH _- -- C Y C L O (HIS-PRO)

k.._!<:

"

80 60

II o

=E 40

1

_O 20 O 'a" .

co

0

I

I

3x10 -e

10 -8

I

10-7

I

10-8

I

10 -s

I

10 -4

I

5xi0

-4

LOG MOLAR CONCENTRATION OF COMPETITOR

FIG. 3 Ligand speeifieity of rat PC/A membrane TRH receptors. Memuranes were incubated in duplicate or triplicate for 6 hours at 0°C with 6.5 nM [3H]MeTRH in the absence and presence of increasing molar excesses of unlabeled competitors. Incubations were t e r m i n a t e d by vacuum filtration. Non-specific binding was determined in each experiment by the inclusion of triplicate membrane incubates containing 6.5 nM [3H]MeTRH and a 1000-fold excess of the unlabeled hormone. Results are tne means and standard errors of 3 to 6 separate experiments. Similar studies were performed on spinal eoro membranes with TRH and analogs (Table 2; Figure 4). Despite the high concentration of potentially interfering low affinity sites in these membranes, the results obtained were very comparable to those for PC/A membrane preparations (correlation coefficient of Ki values between the two tissues = 0.98, p > 0.4). The TRH analog DN-1417, which is equipotent with TRH in causing excitation of ventral motor neurons (8), nevertheless bound only weakly to the TRH receptor in P C / A and c o r d membranes (Ki in the mieromolar range, Table 2). Additionally, the 5-HT antagonist eyproneptadine which has been reported to abolish TRH-evoked motor neuron activity in spinal rats (9), was completely ineffective in preventing specific [3H]MeTRH binding to the TRH receptor in brain and spinal cord membranes (Figure 4). Other, more specific 5-HT antagonists (ketanserin, methysergide) were equally without effect. Specific [3H]MeTRH binding to PC/A TRH receptors was reduced by only 4.6 _+ 2.196 in the presenee of 13 ]aM DN-i417 from pbosplmte buffer stocks but was reduced 44.3 _+7.3% by an identical concentration of the analog using stocks prepared in distilled water (means and standard deviations of four separate determinations; p < 0.02 by Student's "t" test). Because we have found that the TRH analog DN-1417 loses its competitive effectiveness when stored for two weei~s in phosphate buffer at 2°C, all experiments reported here have used solutions of compoundsmade freshly (not more than 24 hours before assay) in distilled water and diluted with buffer immediately prior to incubation with membrane preparations. Others have noted that DN-1417 is unstable in aqueous media, the principal hydrolysis product being "DN-COOH" (20).

608

Analog Specificity of CNS TRH Receptors

e-.-e Me-TRH TRH DN1417

Vol. 36, No. 6, 1985

0 - - o RX77368 41,-- 41, C Y P R O H E P T A D I N E -- -- C Y C L O ( H I S - P R O )

~vlO0 Z

c~

80

1-

6O

Z

I

40 T

20 ,a,.

0

I

3x10 -°

I

I

I

I

10 -8 10-7 10-e 10-s LOG MOLAR CONCENTRATION OF COMPETITOR

I

10-4

FIG. 4 Ligand specificity of rat spinal-cord membrane TRH receptors. Experiments were performed as described in the legend to Figure 3. Results are the means and standard errors of 3 to 7 separate determinations. Search for a specific CNS cyclo(His-Pro) receptor. Our competition analyses demonstrated that the TRH metabolite cyclo(His-Pro) has negligible affinity for the CNS TRH receptor. Additionally, using [3H]cyclo(His-Pro) and methods that should have enabled us to ioentify intermediate to high-affinity binding sites for this ligand, we were consistently unable to detect a specific cyclo(His-Pro) receptor in membranes and cytosols prepared from extrahypothalamic brain, hypothalamus and spinal cord. DISCUSSION Our data present new information regarding affinities of TRH, TRH-analogs and other possibly relevant molecules for the TRH receptor in normal rat spinal-cord and brain. Especiallyinteresting were: a) lack of high-affinity bindingof the TRH-analog DN-1417 by spinal-cord and brain TRH receptor, despite its strong TRH-like action pnysiologieaUyon the CNS, including on lower motor neurons (LMNs) (8); b) lack of highaffinity binding of the TRH-product cyclo(His-Pro) by spinal-cord and brain TRH receptor despite its having some strong TRH-like physiologic actions on the CNS (14,15); and c) lack of any identifiable high-affinity receptor for cyclo(His-Pro) in spinal cord and Drain. These data are important in relation to the beneficial clinical actions of TRH on patients with various motor neuron disorders and have led to a new hypothesis of the method of acute transmitter-like action of DN-1417, TRH, and possibly other TRHanalogs and products on LMNs. We have demonstrated a nigh-affinity, ligand-specific, saturable receptor for TRH in rat brain and spinal cord with properties similar to those previously described (18-21). Until our present studies, we assumed that the spinal cord TRH receptor mediated the acute transmitter-like physiologic effects of TRH on LMN function described in patients (1-3) and animals (5-10). Like TRH, the TRH-analogs DN-1417 (8) and RX 77388 (13) rapidly depolarize LMNs. For example, in chronic spinal rats, DN-1417 produced depolarization similar to that of equimolar TRH, and of somewhat longer duration; DN1417 and TRH were ttlought to be acting directly on the LMN because their excitatory effects were not blocked by tetrodotoxin or baclophen (8). Other animal studies have also suggested that TRH acts directly on LMNs, perhaps by facilitating glutamate ~7,9,10) and/or aspartate excitation. Regardingaffinity of TRH analogs, we found both

Vol. 36, No. 6, 1985

Analog Specificity of CNS TRH Receptors

609

RX 77368 and MK-771 behaved similarly, with Ki values for spinal cord and brain TRH receptor in the nanamolar range, consistent with a TRH-receptor mediated mode of their Known, similar degree of CNS pharmacologic action (27). Surprisingly, affinity of the TRH receptor for DN-1417 was only in the micromolar range in both spinal cord and brain, 2 orders of magnitude lower than that for TRH (Table 2). T h i s result is inconsistent with the idea of high-affinity TRH-receptor mediated DN-1417 excitation of LMNs, which physiologically resembles excitation by TRH itself. Other existing data do not explain this apparent discrepancy. Since DN-1417 and TRH are apparently eliminated from plasma at similar rates (28), the equi-poteney of DN-14i7 is unlikely to be the result of a prolonged retention in the circulation (and thus an enhanced CNS availability) thereby compensating for a much lower affinity. In neuroDenavioral and arousal tests, DN-1417 is as much as six times more potent than TRH (28,29), although its endocrine activity (TSH release in rats) is only 1/40 that of TRH (29). This separation of GNS and endocrine activities of DN-1417 was tnougnt to correlate well with its poor binding to the pituitary TRH receptor (30). However, the dissociation of DN-1417 actions remains unexplainea since we found that brain ana spinal cord TRH receptors also have low affinities for DN-1417 (Table 2), similar to those of the pituitary TRH receptor (30). Our data raise the possibility that DN-1417 may be exciting LMNs by other than a TRH receptor, and by analogy, the action of TRH itself on LIViNs may "also involve that putative other mechanism. To be considered are amine and amino acid neurotransmitter receptors. The receptors for 5-hydroxytryptamine (5 HT, serotonin) are of particular interest. Recently, Barbeau and Beldard (9) reported that electromyographic activity in the quadriceps femoris muscle of chronic spinal rats was rapidly increase(1 by TRH and that the effect was rapidly blocked by cyproheptadine, a 5-HT antagonist. It seems unlikely that this blockade occurred competitively at the level of tne TRH receptor, since we found that micromolar concentrations of cyproheptadine did not block [3H]MeTRH bindingto the TRH receptor in rat spinal cord (Fig. 4) and brain (data not shown). Moreover, we found other 5-HT antagonists (methysergide, spiperone, ketanserin) similarly failed to reduce specific [3H]MeTRH binding to the brain TRH receptor. Since cyproneptadine is considered to act directly at the level of a 5-HT receptor (31), it is possible that DN-1417 and, by analogy, perhaps TRH itself, exert neurotransmitter-like effects at a 5-HT receptor. Interestingly, immunoreactive TRH and 5-HT have been demonstrated in the same medullary raphe neuron somas, axons of which terminate adjacent to LMNs, which are located in the ventral horns of the spinal cord (11,32), and 5-HT receptor activity in the spinal cord is reportedly highest in ventral region (33). Moreover, 5-HT facilitates rat spinal motor neuron excitation by a mechanlsm believed to be similar to that of TRH (I0). W~ile Ono and Fm
610

Analog Specificity of CNS TRH Receptors

Vol. 36, No. 6, 1985

Pyroglutamate and TRH-OH are first-products of TRH. Lectins, by their preferential binding to certain sugar molecules, can help indicate the membrane composition at or near active sites of a receptor. Clinically, quinine, like TRH (3) can markedly reduce painful muscle cramping in patients with various motor neuron disorders. If the CNS TRH receptor is not involved in the transmitter-like actions of TRH, what does it do? We propose that it may moaulate trophic-like influences of TRH on LMNs, as suggested in a study of tissue-cultured motor neurons demonstrating a 17-fold increase of choline acetyltransferase, a transmitter-synthesizing enzyme characteristic of LMNs (12). In pituitary mammatrope and thyrotrope cell systems, TRH also has trophic-liKeactions, e.g., increasing messenger RNA for prolactin (37). An understanding of the mechanisms by which TRH exerts its various influences on LMNs is important to help guide design and selection of TRH analogs to achieve: a) longer action; b) nigher efficacy:side-effect ratio; c) oral effectiveness; d) elimination of the autorefractory state; and e) separation of those analogs more effectively counteracting LMN-cansed weakness f r o m others m o r e precisely counteracting spasticity. In particular it may prove possible to identify analogs having a prolonged trophic benefit to arrest and improve certain human degenerative motor neuron aisorders. ACI~NOWLEDGMENTS Ms. Aloma Ratnasoma provided tect~nical assistance for portions of tnis work. Supported in part by NIH BiomedicalResearch Support Grant #2 S07 RR0 5356-22. REFERENCES I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

W.K.ENGEL, T. SIDDIQUE and J.T. NICOLOFF, Lancet ii 73-75 (1983). W.K.ENGEL, J.R. CARLO,P. VAN DEN BERGH, V. ASKANASand S. DIETRICH, Neurology 34 (suppl I), 238, (1984). W.K.ENGEL, P. VAN DEN BERGH and V. ASKANAS,Ann. Neurol. 16, 109-110 (1984). W.K.ENGEL and T. SIDDIQUE, Neurology34 (suppl 1), 147 (1984). R . A .NICOLL, Nature 265, 242-243 (1977). B.R.COOPER and C.E. BOYER, Neuropharmacol.17, 153-156 (1978). J . W .PHYLLIS and J.R. KIRKPATRICK,Can. J. Physiol. Pharmacol. 57, 887-899 (1979). H. ONO and H. FUKUDA, Neuropharmacol.21, 739-744 (1982). H. BARBEAU and P. BEDARD, Neuropharmacol.20, 477-481 (1981). S.R.WHITE, Society for Neuroscience Abstracts 9, 209.16 (1983). P.C. EMSON, J.M. LUNDBERG and R~F.T. GILBERT, Cell Surface Receptors, (P.G. Strange, ed), pp. 110-125. Jotm Wileyand Sons, New York, (1983). K.M.SCHMIDT-ACHERT, V. ASKANASand W.K. ENGEL, J. Neurochem.43, 586589 (1984). K.A.CLARK and G. STIRK, Brit. J. Pharmac. 80, 561-565 (1983). A. PETERKOFSKY, F. BATTAINI, Y. KOCH, Y. TAKAHARA and P. DANNIES, Mol. Cell. i~iochem 42, 45-63 (1982). V.A.D. WEBSTER and E.C. GRIFFITHS,Thyrotropin-Releasing Hormone, (E.C. GRIFFITHS and G.W. BENNETT, eds), pp. 95-107. Raven Press, New York (1983). D.R.BURT and R.L. TAYLOR, Endocrinology106, 1416-1423 (1980). R~L.TAYLOR and D.R. BURT, Brain Res. 218, 207-217 (1981). N. OGAWA,Y. YAMAWAKI,H. KURODA, T. OFUJI, E. ITOGA and S. KITO,Brain Res. 205, 169-174 (1981). S.M.SIMASKOand A. HORITA, Life Sci. 30, 1793-1799 (1982). S. SHIODAand Y. NAKAI,Cell Tissue Res. 228, 47 5-487 (1983). N.A.SHARIF and D.R. BURT, Brain Res. 270, 2 59-263 (1983). W. ZEMAN and J.R.M. INNES, Craigie's Neuroanatomy of the Rat, pp. 21-25, Academic Press, New York (1963).

Vol. 36, No. 6, 1985

23. 24. 25. 26. 27. 28. 29.

30. 31. 32. 33. 34. 35. 36. 37.

Analog Specificity of CNS TRH Receptors

611

G. SCATCHARD, Ann NY Aead Sci. 51, 660-672 (1949). G.C. CHAMNESS and WJ~. McGUIRE, Steroids 26, 538-542 (197 5). O.H. LOWRY, N J . ROSEBOROUGH, AJ~. FARR and R J . RANDALL, J. Biol. Chem. 193, 265-275 (1951). K. SHIOTA, A. MIYAKE, C. TSUNODA, Y. OKA, Y. NAGAWA and R. NAKAYAMA, J. Immunol. Methods 51, 69-80 (1982). G. METCALF, Brain Res. Rev. 4, 389-408 (1982). M. MIYAMOTO, N. FUKUDA, S. NAURMI, Y. NAGAI, Y. SAJI and Y. NAGAWA, Life Sci. 28, 881-869 (1981). N. FUKUDA, O. NISHIMURA, M. SHITAKA, C. HATANAKA, M. MIYAMOTO, Y. SAJI, R. NAKAYAMA, M. FUJINO and Y. NAGAWA, Chem. Pharm. Bull. 28, 16671672 (1980). (1982). S J . PEROUTKA, R.M. LEBOVITZ andS.H. SNYDER, Science 212, 827-829 (1981). O. JOHANSSON, T. HOKFELT, B. PERNOW~ S.L. JEFFCOATE, N. WHITE, H.W.M. STEINSBUSCH, A.A~I. VERHOFSTAD, P.C. EMSON and E. SPINDEL, Neuroscience 6, 1857-1881 (1981). J.E. LEYSEN, P. VAN GOMPEL, M. VERWIMP and C.J.E. NIEMEGEERS, Adv. Biochem. Psychopharm. 37, 373-383 (1983). W.K. ENGEL, J.F. WILBER, P. VAN DEN BERGH and T. SIDDIQUE, Neurology 34 (suppl 1), 147 (1984). M. PALKOVITS, Hormonally Active Brain Peptides, (K.W. McKERNS and V. PANTIC, eds), pp.279-306, Plenum Press, New York (1982). I.M.D. JACKSON and R.M. LECHAN, Brain Peptides, (D.T. KRIEGER, J.M. BROWNSTEIN and J J3. MARTIN, eds), pp. 661-685, J. Wiley and Sons, New York (1984). E. POTTER, A.K. NICOLAISEN, E.S. ONG, R.M. EVANS and M.G. ROSENFELD, Proc. Natl. Acad, Sci. USA 78, 6662-6666 (1981).