профилактика заболевания

Терапевтический эффект меланина (ENG №3)

Affinity of Melanin for MPTP
1-Methyl-4-phenyl-1, 2,5, 6-tetrahydropyridine (MPTP) is synthesized in accordance with the method described by Schmidle and Mansfield (1955). Purity and identity are confirmed by thin-layer chromatography and gas chromatography-mass spectrometry.

Melanin from beef eyes is prepared according to
Potts, A.M.,Ophthalmol. 3, 405 (1964). The pigment is finally suspended in distilled water to a concentration of 10 mg (by dry weight) per ml suspension. It has been found that the melanin content of pigment granules from beef eyes is approximately 50% (Larsson et al., supra), which gives a concentration of 5 mg of pure melanin per ml suspension.

Synthetic dopamine melanin is prepared by autooxidation (Lyden et al.,sura). The dopamine melanin suspension (in distilled water) is adjusted to contain 5 mg melanin per ml. Both pigment suspensions are stored at2 C.

The binding of MPTP to the melanins is analyzed as previously described in detail by Lyden etal. sura.

Six and one-half ml of various concentrations of MPTP (5.7;M - 1.2 mM) are mixed with 0.5 ml aliquot portions of melanin suspension. The reaction mixtures are incubated at room temperature for one hour. Reference samples contain distilled water instead of melanin. The mixtures are then centrifuged at 35,000 x g for 10 minutes and the concentration of free MPTP in supernatants is measured spectrophotometrically at 243 nm after appropriate dilution. The uptake of MPTP on melanin is calculated from the differences in concentrations between the supernatants and the reference samples.

From the obtained data, the classes of binding sites, association constants and the binding capacity of the melanins are estimated according to Scatchard, G. et al., J. Am. Chem. Soc. 79, 12 (1957). As the molecular weight of melanin is unknown, the value for the number of binding sites is expressed as mol per mg melanin.

The calculations are based on a melanin content of 2.5 mg per incubation.

MPTP is bound to both isolated beef eye melanin and synthetic dopamine melanin in vitro.

The calculated binding parameters are shown in
Table 2. The association constants (K) are expressed asM-1, and the number of binding sites (n), as) mole per mg melanin.

Binding Parameters for the
Interaction of MPTP with Melanin
Beef-eye melanin n1 = 0.09 K1 = 2.32 x 105
n2 = 0.44 K2 = 1.22 x
Dopamine melanin n1 = 0.08 K1 = 5.82 x 105
n2 = 0.05 K2 = 1.22 x 104
n3 = 0.14 K3 = 7.68 x 102
Curvilinear Scatchard plots are observed for both melanins, which indicates that more than one binding class must be implicated. The data for the binding of
MPTP to beef eye melanin could be fitted by the assumption of two classes of binding sites and to dopamine melanin by three classes of binding sites. Both beef eye and dopamine melanin contained a small number of binding sites (n1) with a high association constant (K1) and a great number of binding sites (n2 and n3, respectively). The concordance between the association constants indicates a binding to identical sites on the two melanins.In addition, a small intermediary binding to dopamine melanin was found (n2) which reflects certain differences in chemical structure between the two melanins -- beef eye melanin is obtained from tyrosine as precursor.

The total binding capacity(sun) of beef eye melanin is 0.53pmol/mg melanin. This is probably due to the higher content of carboxyl groups in beef eye melanin (Nicolaus, R.A., in Melanins, E. Lederer, Ed., Hermann,
Paris (1968)). It is interesting to note that the total binding capacity of MPTP to beef eye melanin is of the same magnitude as that of chlorpromazine and cloroquine (Larsson et al.,sura), two drugs that are known to give melanin related side effects.

Thus, it can be seen that melanin is an effective chelator of MPTP, a neurodegenerative disease-causing substance.

Protection from MPTP
Induced Parkinson's Disease
The ability to protect a mammal from toxin-induced neurodegenerative disease, such as Parkinson's disease, is examined by treating monkeys with MPTP in a state in which MPTP could not bind to melanin.

Thirteen male monkeys (Macaca fascicularis) 5 to 8 years old, weighing 3.5 to 4.8 kg, are studied. Four animals are naive controls. Nine receive daily (0.35 mg per kg) injections of MPTP i.v. for four days. Three animals (M1-3) who received no chloroquine are the untreated controls. Six animals are pretreated with chloroquine (4 mg per kg) intramuscularly; three (S1-3) are pretreated for 12 days, and three (L1-3) for 24 days. All six pretreated animals continue to receive chloroquine injections during MPTP administration and for 10 days following MPTP exposure. The neurological examination evaluates the spontaneous movement, tremor, tone and deep tendon reflexes. A zero score in each category reflects a total loss of spontaneous movement, maximum tremor, maximum increase in tone, or maximum hyperreflexia. Deep tendon reflexes examined are brachial radials, knee jerk, and ankle jerk. Muscle tone tests evaluate protraction-retraction, abductionadduction and flexion-extension in both upper and lower extremities. Tremor is rated on severity and the number of extremities involved. Spontaneous movement is evaluated in the morning over a 30 minute period. The rating scale consists of arbitrary units which are weighted to reflect the disturbances most prominent in
MPTP-treated monkeys. Control values are the maximum (normal) score for each element of the exam and all control animals fell within 10% of the control values.

The sum of scores for the four elements is multiplied by 5.26 to provide a total score with a value of 100 for control animals. The results are shown in Table 3.

Neurological Effects of Chloroquine
on MPTPNeurotoxicitv in Monkevs
Neurological Status
Spontaneous Total
Movement Tremor Tone Reflexes (x5.26)
Control 10 5 2 2 100
MPTP 1 1 0.5 0.5 16
M1 1 1 2 1.5 29
M2 1 0 0.5 0 8
S1 3 4 2 2 58
52 3 3 1 1 42
S3 3 3 2 1 47
L1* 0 2 1.5 1 24
L2 5 5 1 1 63
L3 7 5 2 2 84 * Chloroquine level is 36% of that
in other monkeys monitored.

The behavior of the monkeys receiving MPTP alone is similar to that of monkeys previously exposed to MPTP as described in other reports (Schwartzman, R. et al.,
Brain Res. 358, 137 (1985)). On the fifth day after
MPTP exposure the animals manifest decreased mobility and spontaneous movement, abnormal posture, rigidity of the neck and limbs, increased muscle tone, hyperactive reflexes, tremor of upper extremities, and lack of vocalization (Table 3).

Five of the six monkeys pretreated with chloroquine are partially protected from MPTP-induced parkinsonian clinical symptoms. Of the three monkeys receiving longterm treatment with chloroquine, one animal (L-3) is almost completely protected except for a'slight decrease in spontaneous movement. A second animal in this group (L-2) is also protected from the severe effects of MPTP.

Although it exhibits modest rigidity, the monkey manifests no tremor, moves about the cage freely and vocalizes extensively. However, one animal (L-l) demonstrates motor deficits as severe as monkeys receiving MPTP alone; the reasons for the failure of chloroquine protection are described below. All three animals receiving short-term chloroquine pretreatment are partially protected from MPTP neurotoxicity; although they have some rigidity, all move about the cage readily, eat well, vocalize extensively and show only modest tremor.

To determine dopamine and homovanillic acid (HVA) levels, brain samples (10-20 mg wet weight) are homogenized in 300 1 of 0.4 M perchloric acid and centrifuged at4-C for 10 minutes at 1,000 g. Aliquots of the supernatants are analyzed directly by reverse-phase high performance liquid chromatography (HPLC) on an ODS-3 column (Whatman Chemical Separation) with a Pellosil C8 guard column (Alltech Associates). The mobile phase for this system is acetate-phosphate/methanol (95:5) which includes EDTA and sodium heptanesulphonate as an ionpairing agent (Bioanalytical Systems, Inc.). Detection is done electrochemically on a glassy carbon electrode (Bioanalytical Systems Inc.) at an applied voltage of 0.65 V. Plasma levels of chloroquine are determined by
HPLC with UV detection.Samples are deproteinated with 0.2 vol. of 25% trichloroacetic acid followed by centrifugation at4iC for 30 minutes at 1,000 g. Supernatants are removed, lyophilized overnight and reconstituted with 80Cil of 0.1 M perchloric acid. Samples are analyzed directly using reverse-phase HPLC with a
Whatman ODS-3 column (Whatman) and Pellosil C8 guard column (Alltech). The mobile phase for this system is 40% acetonitrile, 0.1 M sodium phosphate (pH 3.0) with 75 mM perchloric acid. Absorption is detected at 343 nm (Bergqvist, Y. et al., Chromat. 221, 2503 (1985)).

Tyrosine hydroxylase (TH) activity is assayed by the tritium release method of Nagatsu, T. et al.,Analvt.

Biochem. 9, 122 (1964) and Levine, R. et al., Analyt.

Biochem. 143, 205 (1984), employing modified reaction conditions of Coyle, J. Biochem. Pharmac. 21, 1935 (1972). Supernatant fluid (50p1) from brain homogenate (1 g tissue in 20 vol 50 mM Tris, H 7.4) are added to 7 ml glass scintillation vials containing 5p1 6 DL-6methyl-5,6,7,8-tetrahydropterine (2.8 mg ml-1), 5 l FeSO4 (2.78 mg ml -1) and 1ILC of ring labelled [ H]tyrosine.

Mixtures are incubated 30 minutes at 37 C and the reaction terminated by adding 501 3 M Na2CO3, pH 11.6.

Toluene/isoamyl alcohol scintillant (5 ml) is then added directly to the vial and the contents mixed for 10 seconds. The results are shown in Table 4. The aqueous and organic phases are allowed to separate and the 3H2O extracted into the organic phase determined.

Biochemical Effects of Chloroquine
on MPTP Neurotoxicity in Monkeys
in plasma DA HVA HVA/DA TH (%
(na ml) (nmol perp tissues control)
C1 47.0 43.5 0.9
C2 42.1 12.1 0.3
C3 73.4 34.6 0.5
C4 64.1 36.2 0.6
M1 1.2 5.6 4.7 11.8
M2 1.4 2.6 1.8 10.8
M3 1.8 5.8 3.2 21.0
Chloroquine and MPTP
S1 ND* 5.2 10.1 1.9 20.0
S2 ND 5.0 8.6 1.7 23.3
83 300 11.2 9.8 0.9 29.8
Chloroquine and MPTP
L1 120 1.1 5.2 4.7 7.2
L2 310 16.0 20.1 1.3 58.5
L3 370 29.6 18.3 0.6 101.0 * Not determined.

TABLE 4 - (Continued)
(nmol perp tissue) control)
C1 62.4 43.8 0.7
C2 39.1 17.4 0.4
C3 79.1 17.3 0.2
C4 53.5 29.9 0.6
M1 0.6 4.1 7.5 6.9
M2 0.6 1.7 2.8 8.4
M3 0.9 8.7 9.4 17.9
Chloroquine and MPTP
S1 1.1 8.0 7.5 14.0
52 1.3 10.5 8.1 18.7
S3 7.6 9.2 1.2 22.6
Chloroquine and MPTP
L1 0.5 5.7 11.1 8.1
L2 11.0 15.7 1.4 39.5
L3 17.3 5.9 0.3 77.1
Results of neurochemical analyses closely parallel the clinical findings (Table 4). In monkeys receiving
MPTP alone, amounts of dopamine, homovanillic acid (HVA) and tyrosine hydroxylase activity (TH) are markedly reduced in both the caudate and putamen. Dopamine is depleted to about 1% of control, whereas HVA was at about 10% of control. The resulting increased HVA/dopamine ratio in the MPTP animals presumably reflects the greater turnover of dopamine in residual dopamine neurons. In monkeys receiving MPTP alone, TH immunocytochemical preparations (Kitt, C.A. et al., Neuroscience 17, 1089 (1986)) reveals a pronounced reduction in the density of TH immunoreactive fibres and terminals in the putamen and to a lesser extent in the caudate nucleus as compared to controls (data not shown). The five chloroquine pretreated monkeys, which are clinically protected from MPTP neurotoxicity, show much slighter reductions in levels of dopamine and TH, as well as TH-immunoreactive fibres and terminals, than
MPTP-treated animals.

Neuropathological findings fit well with the clinical and neurochemical observations. Representative neuromelanin-stained sections through the substantia nigra from each animal are rank-ordered for cell loss by two naive observers whoseranks'nags are identical and closely parallel the neurochemical and clinical results.

The correlation coefficient of the ranking with the caudate dopamine values (R) is 0.90. The greatest reduction in nigral cell number occurs in animals given
MPTP alone (data not shown). Chloroquine-pretreated animals have more surviving neuromelanin-containing neurons, with thegreatest number of cells remaining in the long-term pretreatment group.

Thus, short-term treatment with chloroquine provides partial protection against clinical, neurochemical and neuropathologic effects of MPTP, and in two of three animals, long-term treatment provides more pronounced protection. Why one of the monkeys receiving long-term chloroquine treatment (L-1) is not protected against the effects of MPTP is not known. Chloroquine in plasma is assayed in four of the monkeys immediately before administration of MPTP. MonkeysS-3, L-2 and L-3, which are protected against the effects of MPTP, have 300, 310 and 370 ngml-l chloroquine respectively( 1 ILM of which half is bound to plasma protein).In contrast, monkey
L-l, which developed a parkinsonian syndrome, has a plasma level of 120 ngml-l. Presumably, the failure of drug protection results from the diminished availability of chloroquine in this monkey.

The partial protection of monkeys from MPTP neurotoxicity elicited by chloroquine, together with the high-affinity interactions of MPP+ with neuromelanin (D'Amato, R.J. metal., Science 231, 987 (1986); D'Amato,
R.J. et al., Neurochem. 48, 653 (1987)) indicates that destruction of dopamine neurons in the substantia nigra by exposure to low doses of MPTP is dependent upon interactions of MPP+ with neuromelanin. By inhibiting the binding of MPP+ to neuromelanin, chloroquine may reduce intraneuronal sequestration of MPP+, resulting in reduced toxicity to organelles such as mitochondria (Nickles, W.J. et al., Life Sci. 36, 2503 (1985)).

Example 1 demonstrates that melanin is capable of binding MPTP, a toxin which causes a neurodegenerative disease. Example 2 shows that the disease is caused by the binding of MPTP to melanin in the brain. Since melanin is capable of binding MPTP, it is evident that melanin which is administered to a mammal will bind an environmental neurotoxin such as MPTP, thus preventing a neurodegenerative disease such as Parkinson's disease.

Melanin Administration
to Aid NeuronRecoverv
Male C57BL/6J IMR mice 6-8 weeks of age are used throughout except in one experiment (see below) in which
CB6F1((BALB/cByJ IMR x C57BL/gJ IMR)F1] mice of a similar age are used. Mice are housed five per cage in plexiglass cages with free access to food and water in a colony room maintained at 23 +1 C. Fluorescent lighting in the room is automatically turned on at 06.00 hours and off at 18.00 hours.

[3H]DA (31.6 Ci/mmol) and [3H]mazindol (19.6 Ci/ mmol) are purchased from New England Nuclear (Boston,
Massachusetts). MPTP is purchased from the Aldrich
Chemical Company (Milwaukee, Wisconsin) and converted to the hydrochloride salt as described in Irwin, I. et al.,Neurolosv 35, 619 (1985). Pargyline hydrochloride is a gift from Abbott Laboratories (Chicago, Illinois).

Silver nitrate is purchased from Fisher Scientific Co.

(Fairlawn, New Jersey). All other compounds are purchased from Sigma Chemical Co. (St. Louis, Missouri).

C57 black mice are administered MPTP hydrochloride intraperitoneally according to either one of two schedules: (1) 30 mg/kg/day for 10 days, or (2) 20 mg/ kg/hour for 4 hours. The one group of CB6F1 mice used in this study is administered MPTP according to the following schedule: 50 mg/kg/day for 13 days. This group is used only for anatomical studies looking for cell loss in the substantive nigra cells (SNc). All other studies are performed in C57 black mice.

MPTP hydrocloride is dissolved in distilled water at a concentration such that it could be injected at a desired dosage on a 1 ml/100 g body weight basis. Dose is expressed as the free base.

Melanin is isolated fromStrentococcus antibioticus. Melanin is administered to the test mice at a dose of 10 mg/kg/day following the MPTP treatment by injection into the cerebrospinal fluid until the mice are killed.

The mouse striatum is obtained by placing the brain on its dorsal surface and making two coronal cuts; the first at the caudal end of the olfactory bulbs, the second at the level of the optic chiasma. After placing the resulting brain slice on its rostral surface, one horizontal cut is made just below the corpus callosum and another just above the anterior commissure. Remaining parietotemporal cortex is trimmed away using the external capsule as a landmark. Septal tissue lying between the caudate nuclei is removed by cutting along the tissue planes created by the frontal horns of the lateral ventricles. Striatal tissue thus isolated weighs approximately 20 mg per animal. Immediately after dissection, tissue is wrapped in aluminum foil and stored in liquid nitrogen until assay, with the exception of tissue for uptake studies which is used immediately.

The striatum is weighed, placed in a tube containing 1 ml of 0.4 normal perchloric acid, then homogenized with a Beckman polytron at a setting of 5 for 10 seconds. The homogenate is centrifuged at approximately 20,000 x g for 15 minutes. Concentrations of dopamine (DA), DOPAC and HVA in the supernatant are determined by reverse-phase liquid chromatography coupled with electrochemical detection according to the method of
Mayer, G.S. et al., J.Chromatopr. 255, 533 (1983), with minor modifications. The mobile phase is prepared by mixing 965 ml of 0.15 M monochloroacetic acid with 35 ml of acetonitrile and adding 193 mg of sodium octyl sulfate. This solution is filtered and degassed and then 18 ml of tetrahydrofuran are added.Using this mobile phase at a flow rate of 1.3 ml/min, DA, DOPAC and HVA are resolved using a 4.6 mm x 25 cm C-18, 5IL column (Brownless Labs). Detection and quantitation are performed using a dual series electrode detector (Coulochem Model 5100, Environmental Systems Associates,
Wiggins, Mass.). Electrode potentials are set at +0.4
V (electrode 1) and -0.3 V (electrode 2). The response on electrode 2 ismonitored (10 mV strip chart recorder) and used for quantitation relative to peak heights of known amounts of standards.

The in vitro accumulation oft3H]DA by crude striatal synaptosomal suspensions is measured using the method of Snyder, S.H. et al.,J.Pharmacol.Exo.Ther.

165, 78 (1968) with minor modifications. Briefly, crude synaptosomal suspensions are prepared by homogenizing striatal tissue in 50 vols. (w/v) of ice-cold 0.32 M sucrose, then centrifuging the homogenate for 10 minutes at 1,000 x g. Aliquots (0.1 ml) of the supernatant are added over ice to tubes containing 1.9 ml of Krebs
Ringer phosphate buffer which contained these in final concentrations: 118 mM NaCl, 16.2 mMNa2HPO4, 4.7 mMKC1, 1.3 mM CaCl2, 1.2 mM MgSO4, 1.1 mM ascorbic acid, 11.1 mM glucose, 1.3 mM EDTA, < 0.125 mM pargyline along with equimolar mixtures of(3H]DA and native DA at concentrations ranging from 0.025 to 0.5AM. After vortexing, tubes (except temperature blanks) are incubated in a waterbath at37 C for five minutes, then returned to ice. Synaptosomes are harvested by filtration. Filters are rinsed twice with 5 ml aliquots of physiological saline. Radioactivity in the filters is quantified by liquid scintillation spectroscopy. Assays are performed in sextuplicate at each DA concentration, with half of the samples serving as blanks.Active uptake is defined as the difference between(3HJDA (PM/mg tissue/5 min.) incubation at37 C after correction for uptake at0-4'C.

The binding of(3H]mazindol to striatal membranes is measured according to the method of Javitch, J.A. et al., Eur.J.Pharmacol. 90, 461 (1983).

Nerve terminal degeneration studies are performed using the method of Fink, R.P. et al., Brain Research 4, 369(1967) (Procedure 1). This method makes possible selective silver impregnation of degenerating nerve fibers and terminals. Mice for these studies are killed under sodium pentobarbital anesthesia (40 mg/kg) by transcardial perfusion with 10% formal saline. The brain is immediately removed and stored in perfusion fluid at0-4'C for at least one week before being sectioned on a freezing microtome. 30Mm coronal sections are collected in5% formal saline, then stained with silver according to Fink et al., supra.Mice for these studies are killed one and three days after treatment with either 20 mg/kg/h x 4 or 30 mg/kg/day x 10 of MPTP (n = 3 for each group) or 10 or 20 days after treatment with melanin.

Cell bodies in the SNc are examined in both frozen and paraffin embedded sections after fixation in10% formal saline. Frozen sections (30Am) are stained with silver according to Fink et al., supra Mice for those studies are treated with either 20 mg/kg/h x 4 or 30 mg/kg/day x 10 of MPTP and killed one or three days after the last MPTP injection, at various intervals after treatment with melanin is initiated. Alternating serial paraffin sections (8Am) through the entire SNc are stained with either hemotoxylineosin or luxol fast blue-cresyl violet. C57 black mice used in these studies are treated with 30 mg/kg/day x 10 of MPTP and killed 10 days after the last drug injection. CB6F1 mice used in these studies are treated with 50 mg/kg/day x 13 and killed 21 days after the last drug injection.

The results which are obtained following MPTP treatment and melanin treatment after halting the MPTP treatment are discussed below.

Mice administered 30 mg/kg/day x 10 of MPTP and killed one week later show a 67% reduction in striatal
DA content (Table 1). This result agrees well with that of Heikkila,R.E. et al., Nature 311, 467 (1984). Mice administered 20 mg/kg/h x 4 of MPTP show a comparable depletion of striatal DA (Table 5). The long-lasting depletion of DA induced by this shorter MPTP regimen is dose-related. No lethality is produced by the 2.5, 5 and 10 mg/kg/h x 4 MPTP regimens. Approximately 20% of the mice die after the 20 mg/kg/h x 4 regimen. Larger
MPTP does regimens kill more than 50% of the animals.

One day after cessation of MPTP treatment, mice administered with the 20 mg/kg four-hour MTP regimen or the 30 mg/kg 10 day regimen could not be distinguished behaviorally from their control littermates by casual observation.

Effect of 10-Day and 4-Hour MPTP Treatments
on Mouse Striatal DA Content One Week Later
Treatment n DA ( g/g) % Depletion
Control 10 10.7+ 0.5 -
MPTP 30 mg/
kg/day x 10 5 3.5+ 0.3* 67
MPTP 20 mg/
kg/h x 4 5 2.8 + 0.5* 74 * Significantly different from control group
(P < 0.05; two-tailed Student's t-test).

Along with reduced level of DA, mice treated with 20/mg/kg/h x 4 of MPTP have decreased striatal concentrations of DOPAC and HVA. DOPAC is reduced from 0.96 (+0.14)Ag/g to 0.28(+0.02) g/g and HVA from 1.38(#0.05) Ag/g to 0.60 (+0.06)pg/g (differences significant at 0.05 level). Mice administered 20 mg/kg/h x 4 of
MPTP and killed one week later also show decreased striatal synaptosomalt3H]DA uptake (Table 6). The Vmax was decreased by 62%. TheX; was not changed.

Kinetic Constants of [ H]DA
Uptake One Week after MPTP
Vmax* Km( m)
Control 5540+ 480 0.14+ 0.02
MPTP 2080 + 305 ** 0.12 + 0.02 * Expressed ascpm[3H)DA/mg tissue/5 min.

** Significantly different from control.

The [ H]mazindol binding site has recently been proposed as an additional dopaminergic terminal marker.

Mice administered 20/mg/h x 4 of MPTP and killed three weeks later also show a decreased number of [ H]mazindol binding sites (Table 7). The Bmax was reduced by 44%.

The K4 was unchanged.

Kinetic Constants of [ H]Mazindol Binding
to Striatal Membranes Three Weeks After MPTP
Control 361 17.6
MPTP 20 mg/kg/h x 4 201 17.3 * Expressed as pmol/g tissue.

Three of three mice administered 20 mg/kg/h x 4 of
MPTP and killed one day later for silver degeneration studies show a large amount of fine granular argyrophilic debris in their striata. Some fine granular degeneration is also found in the nucleus accumbens and olfactory tubercle, but in these regions it was much less dense. No such degeneration is found in identically treated sections of control mice, or in other brain regions visible in coronal brain sections at the level of the striatum. None of three mice treated with 20 mg/kg/h x 4 of MPTP but pretreated with 25 mg/kg of pargyline, which blocks the dopaminergic neurochemical deficits induced by MPTP in mice (Heikkila, R.E. et al.,supra show any evidence of striatal terminal degeneration.

In frozen sections through the SNc stained with silver according to the Fink-Heimer method, two of the same three mice which show dense terminal degeneration in their striata show no sign of cell body destruction.

The third animal has a few SNc cells which may have been undergoing degeneration. These few cell bodies stain intensely with silver, appear shrunken, and some have dendritic arbors which were argyrophilic and appear beaded. Cells with similar appearance have been interpreted as undergoing degeneration by various authors.

Although formal counts of these neurons were not performed, affected neurons appear to represent only a very minor fraction of the total SNc cell population.

In serial paraffin sections through the entire length of the SNc, there is no definite cell loss or glial reaction in C57 black mice treated with 30 mg/kg/day x 10 (n = four experimental, two controls) or in CB6F1 mice treated with 50 mg/kg/day x 13 (n = four experimental and four controls) of MPTP. Coded sections from control and experimental animals cannot be distinguished from each other by either of two observers. Mice from these two groups were killed 10 and 21 days, respectively, after drug treatment so as to optimize the possibility of detecting cell loss.

Determination of the level of striatal DA, its metabolites, and synaptosomal uptake at various times after 20 mg/kg/h x 4 of MPTP reveals that substantial recovery in all of these parameters occurs with time.

DA level rises from 28% of control one week after MPTP to 69% of control 15 months later. Three months after
MPTP, there is still a 34% depletion of striated DA.

Partial recovery of striatal DA also occurs after a 30 mg/kg/day x 10 MPTP regimen.[3H)DA uptake capacity likewise recovered with time. TheVmax ofE3H]DA striatal uptake increases from37% of control one week after MPTP to 79% of control three months later (6238 (+ 520) CPM [3H]DA/mg tissue/5 min in control mice vs. 4928 (+ 408)
CPM [3H]DA/mg tissue/5 min in MPTP mice). Over this same time period, DOPAC rises from 29% of control three months later (1.53 +0.09,ug/g in controls vs. 1.03 + 0.03g\g in MPTP mice). HVA rises from 43% of control one week after MPTP (vida supra) to 80% of control three months later (1.36 + 0.11 g/g in control vs. 1.09+ 0.04pg/g in MPTP mice).

When melanin is given following MPTP treatment, the time period required for a similar recovery is reduced and recovery continues through the five-month examination. For example, theVn of[3H]DA striatal uptake increases to 75% of control after 3.5 months of melanin treatment and increases to 85% after five months of melanin treatment.

These results clearly show that during the period tested, melanin is capable of aiding the recovery of neurons following an injury to the neurons. Since melanin is capable of aiding the recovery of neurons following an injury, melanin can be used to treat
Alzheimer's disease.

Any of the melanins isolated in Examples 1-4 may also be administered in the same fashion to aid neuron recovery.

Melanin Treatment of Parkinson's Disease
Male squirrel monkeys (aged 2-3 years) are used for this study. MPTP (Delmar Chemicals) is converted to its hydrochloride salt, dissolved in sterile water to a final concentration of 1 mg/ml (as the free base) and filtered through a 0.22Am millipore filter into sterile injectable vials. All injections are intraperitoneal.

Three different dosage schedules of MPTP are used.

Monkey Group A receives four doses, 2 mg/kg each, which are given at two-hour intervals. Monkey Group B is treated over a five-day period. On day 1, a single 2 mg/kg dose is given. On day 3, two injections of 2 mg/kg each are given, six hours apart. On day 5, a 3 mg/kg dose is given followed by a 0.5 mg/kg dose four hours later (total dose: 9.5 mg/kg). Three doses of 3 mg/kg each are given to Monkey Group C, spaced at sixday intervals. Monkey Group D serves as a control.

After two or more doses of MPTP, increasing bradykinesia and frequent "nodding off" (characterized by closing of the eyes and a slow downward drift of the head) are observed in all animals. Fasciculations of the thigh muscles occur in Monkey Group A. A transient but striking behavioral syndrome is seen after each of the last three doses in Monkey Group A, and after the final two doses in Monkey Group B. This syndrome is characterized by repeated abrupt eye opening and shaking and extension of the extremities.

All monkeys eventually become profoundly akinetic, usually sitting hunched over in a tightly flexed posture. They exhibit a generalized increase in tone.

Vocalization and oral intake were markedly diminished.

Monkeys hold awkward postures for lengthy periods, and sometimes freeze in the middle of a movement. They are often unable to release their grip, getting stuck on the bars of the cage. Tremor and a flexed posture of the arms are seen in Monkey Group C.

Two days after receiving MPTP, one monkey in Group
A is given one-fourth of a 2.5 mg bromocriptine tablet(Parlodel) and one-eighth of a 10/100 carbidopa/L-dopa combination tablet(Sinemete) orally. Within 30 to 60 minutes, the animal is fully mobile and appears almost normal for five hours. A similar response to the same treatment is observed on each of the next four days.

Subsequently, the animal becomes less responsive to medication and is sacrificed on day 10. One monkey in
Group B responds toSinemete (one-eighth of a 10/100 tablet) with full mobility on day 9. However, on subsequent days, he becomes increasingly less responsive to medication, often appearing uncoordinated and shaky.

This monkey is sacrificed on day 15. One monkey in
Group C becomes nearly normal for 24 hours after a single dose of Sinemet (10/100) on day 25. Three days later, this monkey becomes profoundly hypokinetic, develops slow respirations, and dies.

After increased bradykinesis and frequent noddingoff are observed in the monkeys, several monkeys of
Groups A, B and C are administered melanin by injection into the cerebrospinal fluid at a dose of 50 mg/kg daily. The melanin is isolated fromStreDtococcus antibioticus. Amelioration of the bradykinesia and rigidity are seen in the melanin-treated animals. The monkeys' overall functional ability and secondary motor manifestations also improved during the course of the melanin treatment.

Any of the melanins isolated in Examples 1-4 may be administered in the same fashion to treat Parkinson's disease.

Preparation of aMelanin/Boron ComPlex
Melanin analog from Example 4 is solubilized in water at pH 7. The solubilized melanin is then reacted with a 20-fold molar excess of the diazonium salt of 1(4-aminophenyl)-l, 2-dicarba-closo-dodecaborane having a natural abundance of Boron-lO isotope (20%). The resultant melanin/boron complex has about 10,000 boron atoms per molecule of melanin.

This procedure can also be used with the melanin of
Examples 1-4 to yield similar melanin/boron complexes which are soluble in aqueous solution.

Preparation of Monoclonal131I-Anti-CEA IaG
The preparation of monoclonal 131I-anti-CEA IgG is done in accordance with the procedure presented in
Example 2 of United States Patent No. 4,348,376.

Particularly, female, 6-month-old, Balb/C mice are injected with 10-100pg carcinoembryonic antigen (CEA) intraperitoneally, whereby the CEA is mixed in an equal volume (10-1001) of incomplete Freund'sadjuvant.

These injections are repeated one week later, and again two weeks later, but the last injection uses the intravenous route withoutadjutant. Three to four days later, the mice are killed by cervical dislocation. The optimum time for obtaining antibody against a given antigen varies with the antigen, the route of administration, and the timing of immunization, as well as the interval between the last booster injection and the removal of the spleen cells.

The spleens are removed and placed in 60 mm Petri dishes containing either serum-free medium or Dulbecco's
Modified Eagle's Medium (DMEM) with 20% fetal calf serum, at room temperature, and minced with scissors to disperse the cells. The cells are further liberated by agitation for 1-2 minutes on a Vortex mixer. The spleen cells are removed to a conical centrifuge tube and pelleted at 1,000 rpm, the supernatant is removed, the pellet tapped loose, and then resuspended in 5 ml of cold 0.17 NH4C1 for 10 minutes to lyse red blood cells.

Chilled DMEM with 20% fetal calf serum is added and the cells pelleted, and then again suspended in 10 ml DMEM supplemented with 20% fetal calf serum.

The myeloma cell lines used for fusion are maintained in stationary suspension cultures in DMEM with high glucose (4.5 g/L) and 20% fetal calf serum, in 510% CO2 at a cell concentration between 100,000 and 1,000,000 per ml. The myeloma (plasmacytoma) cell lines can be P3/X63-Ag8, which is a Balb/C plasmacytoma derived from MOPC-21(SVasti and Milstein, Biochem. J.

128, 427 (1972)), or a derivative thereof known as FO (Fazekas de St. Groth and Scheidegger, Basle Institute of Immunology, Basle, Switzerland), or 45.6TG1.7, which is a Balb/C line derived from MPC-11 (Margulies et al.,
Cell 8, 405 (1976)). All of these lines lack the enzyme hypoxanthine phosphoribosyl transferase (HPRT; E.C. and are thus killed in a selective medium containing hypoxanthine, aminopterin, and thymidine (HAT), as described by Littlefield (Science 145, 709 (1964)).

The spleen cells obtained from the immunized animal are then fused with the plasmacytoma cells by using polyethylene glycol according to an adaptation of the method of Gelfer et al. (Somatic Cell Genetic. 3: 231236, 1977). For example, a 30% polyethylene glycol solution is made by heating sterile polyethylene glycol 4000 (Merck, molecular weight of about 4,000) (0.5 g
Polyethylene glycol + 0.05 ml dimethyl sulfoxide (DMSO) + 0.5 ml distilled water) and DMEM without serum to41it, and mixing 3 ml of polyethylene glycol with 7 ml
DMEM without serum, pH 7.4-7.6, and kept at37iC until use. Fusions are made at room temperature. The myeloma cells (106-107) are washed twice in serum-free medium and then mixed with 1-3x107 to 1-3x108 spleen cells in 50 ml conical bottom centrifuge tubes (Falcon 2070).The cells are centrifuged at 250xg for 5 minutes and the supernatant fluid is carefully aspirated. A 0.2 ml amount of the polyethylene glycol preparation is added, and the tube is gently agitated by hand to resuspend the cells. Next, the cells are centrifuged for 3 minutes at 250xg and again at 400xg for another 3 minutes, and then kept undisturbed for an additional 3 minutes. The cells are exposed to polyethylene glycol for about 8 minutes.

Thereafter, about 5 ml of serum-free medium is added to the tube, the cells are resuspended gently, and then repelleted by centrifugation at 250xg for 5 minutes.

The supernatant is removed and the cells are suspended in 20 ml of serum-containing medium and incubated at37 C, in a humidified incubator for 48 hour before being placed in microplates to which HAT medium is added.

Alternatively, the cells are immediately suspended in 30 ml of a medium consisting of DMEM, 10% NCTC 109 medium (Microbiological Associates), 20% fetal calf serum (GIBCO), 0.2 units bovine insulin/ml (Sigma), 0.45 mM pyruvate, 1 mM oxaloacetate, and antibiotics of choice.

Thymidine(1.6x10-10 M) and hypoxanthine(1x10'4 M) are added. The cells in this medium are distributed into 6 microplates (Linbro FB 96 TC) with 1 drop (about 501) per well. The next day 1 drop of the above-specified medium containing thymidine and hypoxanthine, now with aminopterin (8x10-7 M), is added to each well. Two drops of the medium of above is added 6-7 days later and clones appear microscopically between 10 and 20 days.

The hypoxanthine-aminopterin-thymidine (HAT) medium can also be added immediately after the fusion, or at a later time. An improvement in the number of hybrids obtained is made when a feeder layer is added to each microwell. Here, human fetal fibroblasts are irradiated with 4500 r, and 1,000-2,000 such cells are added to each well, either the day before the fusion or directly to the fused cells and so dispensed with them into the microwells. After clones have appeared macroscopically, the medium is changed by removing most of the medium and adding fresh medium. After a second change of medium, the medium is left for at least 4 days and then collected for assays of antibody activity and specificity by conventional assays.

Large amounts of antibody are obtained from spent culture medium harvested from 150 mm plates or roller bottles. The medium is subsequently concentrated by means of a hollow-fiber concentrator (Amicon). Also, antibody is obtained from the ascites fluid of athymic (nude) mice (nu/nu) that were injected 2-3 weeks previously with about 1 billion cloned hybridoma cells.

The ascites fluid is diluted with saline by flushing the peritoneal cavity of each mouse with saline, the diluted fluids from each mouse are pooled.

The monoclonal anti-CEA IgG is radiolabeled with I131 by injection into a radionuclide vial containing131I (Amersham-Searle). The monoclonal anti-CEA IgG is injected at a concentration of 20Mg IgG per mCi131I.

Preparation of A Melanin/Boron
The monoclonal anti-CEA IgG of Example 10 is reacted with a three-fold molar excess of the melanin/boron complex of Example 9 in an aqueous solution at pH 8.

The reaction proceeds overnight at aboutO C. The resultant conjugate is separated from unreacted melanin/ boron complex by passage through a sizing column. The recovered conjugate is then stored as a sterile solution.

Any of the melanin/boron complexes of Example 9 may be used in the above reaction to form a melanin/boron antibody conjugate.

Neutron CaptureTherapy of Tumors
A patient having a cervical cancer is injected with 0.9 mg of the melanin/boron antibody conjugate of
Example 11. The injection is preferably in the form of three injections of 0.3 mg of conjugate spaced about 3-6 hours apart for a 1 gram tumor.

A collimated beam of thermal neutrons is focused on the tumor location, and the tumor is irradiated with an external neutron beam dose of 400-800 rads delivered over an 8-20 minute period. Optionally, the procedure may be repeated, but usually should not exceed a total dose of 3200 rads.

Any of the melanin/boron antibody conjugates of
Example 11 may be used in the above described neutron capture therapy procedure.

Preparation of Cloned HumanTvrosinase
Cloned human tyrosinase is prepared using the method of Kwon, B.S. as described in the published PCT application WO 88/02372.

The tyrosinase is produced in E. coli strain
MM 294. The Amel 34 cDNA (as described by Kwon, B.S. in the same PCT application) is fused to a Tac expression vector (U.S. Pharmacia Inc.) which hasTro and lac promotor together. The construct is expressed in the E.

coli strain MM 294 and subsequently purified by affinity column chromatography.

This tyrosinase is then used to treat diseases caused by a melanin deficiency.

Introduction of Human Tyrosinase Gene
Into a Defective HSV-1 Vector
A defective herpes simplex virus 1 (HSV-1) vector, pHSVlac, has been developed by Geller, A.I., et al.,
Science 241, 1667 (1988). This vector is useful for transporting genes through the blood brain barrier.

The vector, pHSVlac, contains the Escherichia coli lacZ gene which is under the control of the HSV-1 immediate early 4/5 promoter. Using conventional publicly available endonucleases, pHSVlac is digested at its
EcoRI sites to remove the E. coli lacZ gene. TheA mel 34 human tyrosinase gene (described by Kwon, B.S. in PCT application WO 88/02372) is then inserted to pHSVlac in place of the E. coli lacZ gene, and the vector is religated using conventional techniques.

This chimeric pHSVlac vector may then be used to introduce the tyrosinase gene into patients suffering from diseases caused by a melanin deficiency.

Stably Transforming Cultured
Peripheral Neurons with the pHSVlac
VectorExpressing Tyrosinase Gene
Primary cultures of dissociated neurons from dorsal root ganglia and superior cervical ganglia of newborn rats are prepared in accordance with the techniques taught by Hawrot, E. et al., Methods Enzvmol. 58, 574 (1979). The cultures are then infected with the chimeric pHSVlac vector of Example 14, above, and incubated for 24 hours at37 C. The cultures are then fixed and assayed for tyrosinase using antityrosinase antibodies (available from Dr. Seymour H. Pomerantz, Department of
Biological Chemistry, University of Maryland School of
Medicine, Baltimore, Maryland 21201) and conventional techniques.Tyrosinase is found to be present in both the dorsal root ganglia cell cultures and the superior cervical ganglia cell cultures.

Transneuronal Transfer of the pHSVlac
VectorExpressing Tyrosinase Gene
In accordance with the technique of Ugolini et al.,
Science 243, 89 (1989), eight rats (6 to 7 weeks old) are unilaterally injected in the ulnar and median nerves with the chimeric pHSVlac vector of Example 14, above.

After four days, the rats are anesthetized and perfused with 10% Formalin as taught by Ugolini et al., Brain
Res. 442, 242 (1987). The brains and spinal cords of the rats are cut into 60ym transverse frozen sections, and the presence of tyrosinase is assayed using antityrosinase antibodies and conventional techniques as described in Example 15, above. Tyrosinase is found to be present in the rat brain neurons, due to the transneuronal transfer of the chimeric pHSVlac vector from its peripheral neuron injection site to the brain.

The tests of Examples 17-19 below were conducted to analyze the neurological affects of the melanin analog produced by the method of Example 4. All of the tests described in the Examples below were completed within three minutes.

Thirty two male Long Evans rats were used in the neurological analyses. The rats were divided into 4 groups of 8 animals in each group. The first group received an injection of 1.0 mg/kg, the second group received an injection of 0.1 mg/kg, the third group received an injection of 0.01 mg/kg, and the last group received a 0.9% saline injection. All injections were administered I.P. 30 minutes prior to behavioral testing. All animals were injected once a day for four consecutive days. The animals were given the following tests in the described order:
Sensorv Tests
A. Olfaction orienting test. Since normal rats will tend to orient to novel stimuli, two pungent (to humans) scents were used in this test (ammonia and
Mennen Skin Bracer) to determine the effects of melanin analog on the olfactory orienting response. A cotton swab was moistened with either scent.Then the swab was brought from behind the animals's head toward the nose.

The swab was not allowed to enter the animal's visual field nor was it allowed to touch the animal. The rating scale of response was 0 if no orienting response occurred and 1 if the animal oriented by sniffing at the swab. There were four tests (2 with each odor) brought from the right and the left sides. The average of the four tests was used as a index of olfactory detection.

B. Visual stimulus orienting test. A 2x2 inch corrugated cardboard square was held by a pair of hemostats. This square was brought from behind the animal's head into its peripheral field of vision. The rating scale for response was the same as above. There were two tests, one from the right and the other from the left. The average of the two tests was used as an index of visual detection.

C. Somatosensorv orienting tests. Using a Von Frey hair of 2 g pressure, the rat's shoulders, mid-section, and hindquarters were touched on both right and left sides. The average of six responses was used as an index for somatosensory detection.

D. Whisker touch orienting test. A cotton swab was brought from behind the animal's head out of the animal's visual field and put in contact with the vibrissae successively on their right and left sides. The response recorded was based on the same scale as tests 1 and 2.

The average of the two responses was used as an index of whisker touch detection.

E. Results. The effects of the various melanin doses on detection of olfactory cues, visual cues, somatosensory cues and whiskey touch were measured. In general, it was found that the melanin analog does not alter sensory function.

Sensorv-Motor Tests
A. Placing reflex test. For this test the animal was suspended by the tail. While suspendedtheanimal was brought close to the edge of a table. If the only stimulus necessary for forelimb extension was the sight of the table, it was rated as a 1, if it required the touch of the snout, the score was 1/2, if it required the maintenance of the snout touching the table, the score was 1/4, and if no response was elicited, a 0 was given.

B. Tilted platform test. Each animal was placed on the center of a 30x30 cm square of carpet-covered plywood. The plywood was tilted down to 30 degrees so that the rat's head was at the low end. If the animal responded normally it would turn around so its head faced up the slope. This response was given a 1. If the animal remained in the original position a0 was given.

C. Results. The effects of the various melanin doses on performance of the placing reflex and the tilt reflex were measured. In general, the melanin analog does not alter sensory-motor function.

Motor Tests
A.Grasping reflex. The rat was suspended by the nape of the neck then the palms of both front feet were touched by a single piece of stiff piano wire. Grasping is accomplished by flexion of the fingers around the wire. The rating scale used was 0 for no response, 1 for grasping the wire.

B. -Riahtina reflex (rollover). In this test the animal was held upside down 40 cm above a foam pad; then the animal was dropped. If the animal landed on the its feet a score of 1 was given, if anything else a 0.

C. Results. The effects of the various melanin doses on performance of the grasping reflex and two types of righting reflexes were measured. In general, it was seen that the melanin analog does not alter motor functions.

Food and Water Intake and Open Field Activity
One week after the completion of the neurological analysis, the animals were placed on a 23 3/4-hour water and 23 1/2-hour food deprivation schedule. Water intake was measured once per day for the 15 minute water drinking period. Food intake was measured once per day for the 30 minute feeding period.

The animals were adjusted to this schedule for 5 days. On the sixth day, 30 minutes prior to the water and food intake period, the same groups of animals received the various melanin analog doses. On the seventh and eighth day, water and food intake were again measured.

The animals were then maintained on the food and water schedule for four more days. On the fifth day, the same groups of animals received the various melanin doses, and 30 minutes later they were tested in an open field. The open field for evaluating level of activity and emotionality consisted of a large wooden square box (120 x 120 cm) with 29 cm high walls. The floor of the box was painted white and divided by black lines into 64 square sections each measuring 15 x 15 cm. Thirty minutes after drug injection, each animal was placed in the center of the open field for a 10 minute period.

The number of squares entered and the number of different activities (grooming, scratching, righting, washing, defecating and urinating) that were emitted during each minute was recorded. A "square entry" in the open field consisted of having all four feet within one square.

The effects of various dose levels on water intake on day 6 were examined. There appears to be a dose dependent decrease in water intake with marked reductions at the higher dose levels. There was total recovery of water intake the day after the melanin analog injection.

The effects of various dose levels on food intake on day 6 was also examined. There was a clear dose dependent reduction in food intake with marked reductions at the higher dose levels. Recovery of food intake to normal levels was seen the next day with the two lower drug levels. With the highest melanin dose, recovery of food intake was not seen until 2 days later.

In general, the melanin analog affects normal food and water intake suggesting an influence on neural systems mediating variables that control level of motivation.

The effects of various melanin analog dose levels on general activity level (number of squares traversed in a 10 minute period) was also studied. Relative to saline or the low dose injection group, there is a marked reduction in overall level of locomotion in animals that received the higher dose levels (0.1 and 1.0 mg/kg). The effects of various melanin dose levels on activities other than locomotion (grooming, righting, scratching and washing) were measured. Relative to saline or the low dose injection group, there is a marked reduction in overall level of activity in animals that received the higher dose levels (0.1 and 1.0 mg/kg). The effects of various melanin analog dose levels on emotionality (defecation and urination) were studied as well. There appears to be some increase in the level of emotionality in the group that received the 0.1 mg/kg dose relative to the other groups.In general, the higher melanin analog doses produce a general inhibitory effect on locomotion and other behavioral activities, again suggesting a drug effect on motivation.

Acauisition of Radial Arm Maze
Due to the motivational effects of the high dose levels of the melanin analog, it was decided not to use the highest dose levels (1.0 mg/kg) for this experiment.

Thus, 24 animals were food deprived and maintained at 80-85% of their ad lib. weight for at least one week.

For the next 6 days all animals were adapted to the radial arm maze with food continuously available in the center and at the end of the arms.

Starting on the seventh day, 30 minutes prior to testing, the animals received either an 0.01 mg/kg (N=8), or 0.1 mg/kg (N=8), or a saline (N=8) injection.

Each animal was placed in the center of the radial arm maze and allowed to visit the 8 arms which were baited with small pieces of Froot Loops cereal. Each animal was run until it had visited and consumed food at all 8 arms or until 15 minutes had elapsed. Reentry into an arm visited previously was scored as an error. The total number of errors to visit all 8 arms was used as the dependent measure. The above procedure was repeated once a day for 12 days.

The results indicate that all three groups learned the maze (showed a reduction in errors) at about the same rate, even though there is more day to day variability in the groups with the melanin analog injections. Also the melanin injected groups took more time to finish each trial. In general, the melanin analog does not appear to facilitate or impair acquisition of the radial arm maze. In conclusion, high levels of melanin analog do not impair sensory motor or motor functions, nor do they alter acquisition of a radial arm maze task, but this agent does alter motivational levels by inhibiting water and food intake and overall levels of activity and locomotion.

Melanin's Effect
On Carotid Nerve Fibers
Melanins were tested for their efffect on carotid nerve fibers. The six substances were: F5H (melanin made in the presence of 5mM Fe precipitated with HC1 at pH2), COH (melanin made in the presence of 1 mM Cu precipitated with HC1 at pH2), F4E (melanin made in the presence of 4 mM Fe precipitated with two volumes of 95% ethanol), COE (melanin made in the presence of 1 mM Cu precipitated in the presence of two volumes of 95% ethanol), C47 (melanin precipitated with hCl at pH 2) and S23 (carborundum). The melanin analogs were tested against their solvent, and dopamine (DA). The solvent was 0.2NH,OH, at pH 7.4 adjusted with 1NHC1, referred to asNH*C1 or solvent.

The biological preparation used was the isolated rat carotid body with its nerve, superfused in vitro with oxygenated Tyrode's solution, pH 7.4 at36 C. The superfusion flow was 1 ml/minute. The carotid body is an organ strategically located in the carotid bifurcation, that senses the changesP 2, pCO2 and pH of the arterial blood going to the brain. The sensory elements are the glomus cells. They are synaptically connected to the carotid nerve fibers which carry the information to the central nervous system. These nerve fibers have dopaminergic receptors which are well identified biochemically and physiologically.

The carotid nerve discharges were recorded with platinum electrodes (0.2 mm diam.) and AC amplified. The nerve discharges were monitored in an oscilloscope and stored on tape for further analysis. Later, the nerve discharges were counted using a window discriminator and a digital counter. The impulse frequency was measured every second and registered on paper using a chart recorder. The data was analyzed using a Macintosh computer.

Some problems developed and were corrected during the experiments:
1) The melanin analogs precipitated when added to the perfusing solution. This problem was avoided by increasing the solvent volume.

2) DA oxydized when dissolved in the solvent.

Thus, it was difficult to test DA again the melanin analogs using the same solvent. Therefore, DA had to be diluted in Tyrode's solution.

3) The solvent had has a stimulating effect on the chemosensory discharges of the carotid nerve due to the ammonium ions. Thus, to measure the potency of the melanin analogs the maximum discharge rate induced by the solvent was subtracted from the maximum discharge rate induced by the melanin analog. The resulting discharge frequency was termed aFc.

The basal spontaneous discharge of the biological preparation was stable along the experiments, and to avoid interaction between the substances being tested, the preparation was washed with superfusing solution for 10 to 15 minutes between melanin analog and control injections.

A. Effect of 100 up Doses of Melanins
All of the samples tested were applied in 101 volumes and they produced discharge stimulation. For all the substances tested the stimulating effect was greater than that induced by the same volume of control solution (solvent). The injection of 10 1 of Tyrode's solution did not change the discharge frequency. COH and COE were the most potent substances, and the amplitude of their peak effects was not significantly different(p < .001). Taking these effects as 100%, the response of the biological preparation to F5H was 27% lower, that of F4E was 54.2% lower and that of C47 and
S23 was 59% lower.

B. Effect of 50 ul Doses of Melanins
In this part of the procedure, all of the samples were applied in 51 volumes. The effects of melanins decreased in amplitude as expected when lowering the dose. However, the application of 50pg of F4E produced a discharge increase61.7% greater than that induced by 100Ag of the same melanin analog. Also, the effect of
C47 was about the same as that obtained with 100ILg of the same melanin analog. These results indicate that the dose response curves for these two melanin analogs should be shifted to the left.

C. Effects of DA and the Melanins
Dopamine produces two different effects on the biological preparation. Low doses (10-30Mg) produce discharge depression followed by stimulation. In contrast, large doses (50-300Crag) produce only discharge stimulation. Although DA could not be applied dissolved in NH4C1, its peak effects can still be compared to those induced by the melanin analogs.

Low doses of F5H had an effect very similar to that of DA. Therefore, it is possible that both F5H and DA may share the same receptor sites. This theory is supported by the finding that F5H is less potent when applied to the bath immediately after DA. Furthermore, when the application of F5H is spaced from DA, the response of the biological preparation to F5H is unaffected. This experiment should be repeated using dopamine recaptor blockers to study this phenomenon in more detail. It is also interesting to note that in several experiments a sharp depression of the nerve discharges was observed during the rising phase of stimulation produced by some melanin analogs.

D. Duration of the effect of Melanins
Another interesting finding is that the effect of the melanin analogs was 30% to 50% longer than that of
DA. Also, the effect of the melanins lasted longer than the effect of the solvent. This could indicate a stronger binding of melanins to the receptor sites or a slower inactivation mechanism for these substances.

There is extensive evidence that the carotid nerve endings have dopamine receptors. The above results show that the carotid body and/or nerve also has receptor sites for melanin analogs. This conclusion is supported by the dose dependent effects reported in the results.

The effects of melanins on the biological preparation are more potent than DA and last 2-3 times longer.

The most potent effects for larger doses (100pg) of melanins were those induced by COE and COH. However, for lower doses (50Ag) F4E induced a very potent discharge stimulation. The smaller effect of F5E at higher doses could be explained by receptor desensitization, a well known phenomenon in synaptic physiology.

In decreasing order of potency would follow COE COH F5H
C47 and 523.

In view of these results, F4E is the most potent substance. An analysis of response curves shows that
F4E and CDE are the most potent with F4E having the strongest effect. This differences in potency is clearly seen at doses of 10Ag at which there is statistically significant difference between F4E and CDE (p < 0.001). At a dose of 25Ag, there is not a statistically significant difference between F4E and CDE.

It is also important to consider that all of the substances were applied to the bath as a bolus, diluted in saline and at a distance of 1 mm from the preparation. Since there was a continuous flow of saline in the recording bath, the actual concentration of the substance of the carotid body would be less, perhaps as much as a50% dilution.

Although it is unclear if melanins and DA share the same receptor sites, the depression of the effect of F5H by previous application of DA is an indicator that this may be true.

While the invention has been disclosed by reference to the details of preferred embodiments of the invention, it is to be understood that the disclosure is intended in an illustrative rather than in a limiting sense, as it is contemplated that modifications will readily occur to those skilled in the art, within the spirit of the invention and the scope of the appended claims.