In 2000, when Arvid Carlsson learned that he was awarded the Noble Prize for Medicine, he told the reporters in his very first statement: “Well, anyway, I had to wait for this for 40 years!” Though the statement was bold, it was also perfectly appropriate: As early as 1957, Carlsson had published a seminal paper in Nature which demonstrated that in rabbits the immobility caused by the treatment with reserpine could be antagonized by levodopa, the precursor of dopamine (DA) (Carlsson et al. 1957). He thereby established DA, so far considered only as a noradrenaline (NA) precursor, as an independent neurotransmitter crucial for motor functions (Carlsson et al. 1997).

Only a few years earlier, Reserpine had been isolated from the Indian plant Rauwolfia (known as a kind of panacea to traditional Indian medicine) and found to have antipsychotic properties. It was subsequently established to have anti-hypertensive effects as well, but also to cause depression and Parkinsonism at higher dosages. These observations formed the basis both for Carlsson’s 5-HT (serotonin) or monoamine deficiency hypothesis of depression and for his DA deficiency hypothesis of Parkinson’s disease (PD). He also developed the DA hyperactivity hypothesis of psychosis (Carlsson 1959). His review on brain monoamines had strong support from biochemical and histochemical detection methods based upon the Falck–Hillarp fluorescence technology (both authors know from own experience how difficult and often unpredictable fluorescence methods can be!). Carlsson developed these methods with the help of Kjell Fuxe and other colleagues from the Karolinska Institute at Stockholm (Carlsson et al. 1962), and thereby founded the famous Swedish school of monoamine pharmacology.

Walther Birkmayer from Vienna soon learned about Carlsson’s levodopa work. He immediately injected his 1st patient with this new compound and, indeed, could achieve a very clear albeit short lasting anti-akinetic effect (Birkmayer and Riederer 1983). This historical pilot study can still be seen in a short movie, now in the hands of Peter Riederer, Birkmayer’s former student.

In a stroke of genius, Walther Birkmayer subsequently added a decarboxylase inhibitor (Birkmayer and Riederer 1983). In this way, he could quite unexpectedly increase and prolong the anti-Parkinson efficacy of levodopa whilst at the same time reducing its side-effects (especially nausea and emesis). In fact there has always been a rumour that Arvid Carlsson, in the good old tradition of pharmacologists, had swallowed a very significant amount of levodopa soon after his discovery, only to experience severe and long-lasting emesis and to conclude that levodopa would never become a useful treatment. In any case Marcus Guggenheim from Hoffman LaRoche, who equally isolated levodopa from beans, experienced severe vomiting after testing it; understandably enough, he decided not to investigate it any further (for a comprehensive review see PB Foley, Beans, Roots and Leaves, Tectum Verlag Marburg 2004).

The introduction of DA agonists into therapy, however, had started much earlier when in 1869 apomorphine was first derived from morphine (itself isolated by Friedrich Sertürner not much earlier, 1805) by cooking morphine with a strong acid such as hydrochloric acid or sulfuric acid (Matthiesen and Wright 1869). Whilst we do not know who detected apomorphine’s emetic effects, which are even stronger than those of l-DOPA, apomorphine became a quite commonly used emetic as early as the 1870s and replaced the much more toxic calomel. The first comprehensive evaluation of apomorphine’s effects was made in Siebert’s 1871 thesis at Dorpat (now Tartu) University as reported in a review paper from Magnus 1920. Of course, its DA agonistic mechanism of action was not yet known.

Due to a first pass effect of 100% when administered orally, apomorphine must be injected to be effective. Emetic dosages are in the range of 1–4 mg, with some variability; in 1875, Jurasz gave 1.5–5.0 mg, an elder people 5–20 mg s.c.as effective dosages, whilst the latter dosage was also be found in the DAB 7 (quoted from Magnus 1920).

At lower dosages apomorphine has a sedative effect, which is why it has been used in patients with agitation (e.g., from alcohol intoxication) when other sedatives were contraindicated. In addition to treating oral intoxications as an emetic, apomorphine has also been used in so-called aversive treatment: as proposed by Douglas (quoted from Magnus 1920), alcoholics got an apomorphine injection simultaneously with an alcoholic beverage to induce an aversion against the alcohol stimulus (in the sense of a conditioned reflex); similar “therapies” have even been used for transvestites and homosexual persons.

J. Y. Dent used apomorphine to treat alcohol addicts, and his method was made famous by W. S. Burroughs’ report in Harper’s magazine (“Kicking drugs: a very personal story”, Harper’s magazine 235, 1406, 39–42, 1967), though this may in part have been due to Dent’s personal charisma; at least Burroughs concluded that “Dr. Dent without apomorphine is better than apomorphine without Dr. Dent!” (Harper’s, ibidem, p. 41).

By the time of Arvid Carlsson, apomorphine was, therefore, just an obsolete emetic. And yet bulbocapnine, an alkaloid from Corydalis bulbosa (spring fumewort or bird in the bush, in German language Lerchensporn) that shows great structural similarity to apomorphine, had been used to treat trembling hands as early as in the sixteenth century (quoted from Goodman-Gilman, Pharmacological Basis of Therapeutics, II edition 1950, p. 213). Long before the use of apomorphine in PD, Rudolph Magnus had written the comprehensive and extended review about apomorphine (Apomorphin, Apocodein und Ipacacanha Alkaloide, in: A. Heffter, Handbuch der Experimentellen Pharmakologie 2/I, 430–453, Springer, Berlin, 1920). Here he reviewed experiments by Harnack (1874) and Feser, who in 1873 had reported that in animals, apomorphine induced typical behavioural syndromes such as “continuous walking and running” in cats and dogs (in the later species interrupted by violent vomiting), but also in pigs (“un-interrupted snuffing and digging in the earth”), cows (“uninterrupted licking of the feeding trough”), horses (“continuous trotting”) and rodents and rabbits, in whom apomorphine caused continuous sniffing, licking, and gnawing, and finally in pigeons, where it caused pecking to the point of exhaustion (Feser 1873). All these are typical frequent movements of these animals which apparently, once started, cannot be stopped. Nowadays we call this stereotyped behaviour as an expression of exaggerated dopaminergic activity.

We do not know why R. S. Schwab applied apomorphine to patients with PD already in 1951 (in the case of amantadine, it was a fortuitous observation of one of his PD patients who took this drug to improve her flu and found her PD improved). In 1953, the German neurosurgeon A. Struppler together with Thure von Uexkull infused apomorphine i.v. to two PD patients under the assumption to achieve a vegetative Umstimmung (a vegetative changeover) and found the tremor greatly improved (Struppler and Von Uexkull 1953). In 1974 F. Strian noted the same improvement in PD symptoms after i.v. apomorphine infusion in a few patients, but stopped his investigations because “the previously inconspicuous farmer’s wife developed a psychosis with sexual connotations” which was definitely unwelcome at the Munich Max-Planck-Institute for Psychiatry.

Since parenteral therapies, especially those which suffer from as short a half-life as apomorphine (t/2 about 15 min) were quite cumbersome, in 1970 George C. Cotzias wanted to establish the anti-PD effective dose of oral apomorphine after having confirmed Birkmayer’s findings with l-DOPA. Cotzias went up to single dosages as high as 150 mg before having to stop these investigations as his PD patients developed a nephropathy—no surprise with apomorphine bioavailable to 100% but also with the 100% first pass effect resulting in toxic quinone metabolites (Rehse 1978). Given that A. M. Ernst had pointed out structural similarities between apomorphine and DA already in 1966, a number of clinicians then successfully used apomorphine in PD patients. However, they only did do on a parenteral basis, especially after K. Fuxe and his colleagues had confirmed the potent dopaminergic activity of apomorphine (Andén et al. 1967). The dopaminergic effect of apomorphine has also been confirmed by its effects on prolactin, as its injection causes a clear-cut, but very short-lasting reduction in normal or elevated prolactin levels. This came as another confirmation of Bob MacLeod’s proposal according to which DA lowers prolactin and so do dopamine agonists (MacLeod and Lehmeyer 1974).

Another classical DA agonist is bromocriptine (CB-154, 2-bromo-ergo-criptine, bromocriptine). This 2-brominated derivative of the natural ergot constituent ergocriptine was a result of attempts from Sandoz AG/Basle (now Novartis) to find new uses for the great amount of derivates of ergot, a true chest of treasures. Sandoz research scientists were thus looking for derivatives with a new pharmacological profile. Bromocriptine did in fact differ from its parent compound as it had, in contrast to ergocriptine, an anti-nidation and thus contraceptive effect in rats (Flückiger and Wagner 1968). Rumour has it that with this compound Sandoz AG had wanted to compete with former Schering AG in Berlin, then the world leader in sexual steroids and thus in contraceptives. It must have come as an unpleasant surprise to their management that bromocriptine had no contraceptive effect whatsoever in humans: the nidation of a fertilized egg in humans does not depend on the presence of pituitary prolactin (quite in contrast to the situation in rats), and cannot, therefore, be inhibited by the DA agonist bromocriptine. Quite on the contrary, bromocriptine in humans was actually found to have pro-fertility effects, at least in people with elevated prolactin levels. This is because in humans, elevated prolactin levels inhibit the pulsatile luteinizing-hormone-releasing hormone (LHRH) secretion from the hypothalamus which would otherwise cause an LH surge to induce steroidogenesis in the ovary or the testes, and thereby inhibit oestradiol production (and thus of ovulation) or the synthesis of testosterone, which is necessary for spermatogenesis and potency. Hyperprolactinemia can be idiopathic or result from prolactin producing micro-or macro-adenomas of the anterior pituitary. That bromocriptine did exert its prolactin-lowering effect as a DA agonist was proven, once more, by Kjell Fuxe and his co-workers from the Karolinska Institute in Stockholm (Fuxe et al. 1974). Even earlier, Wuttke et al. (1971) have been the first endocrinologists to have shown in the case of the natural ergot derivative ergocornine that these compounds could lower prolactin.

Whilst infertility is obviously as much an individual catastrophe as overpopulation is a problem of all mankind, the Sandoz management must nonetheless have felt rather unhappy, since fertility and its disorders provide a much smaller market than contraception. They, therefore, gladly accepted Donald B. Calne’s offer of studying this new compound in patients with PD, where it eventually turned out to be effective either as monotherapy or in combination with levodopa (Calne et al. 1974).

Thus bromocriptine became the first oral DA agonist in the therapy of PD, despite its association with a rather great number of adverse events (mostly similar to those known for ergocriptine and the other ergot derivatives) and despite the fact that even at high dosages its anti-PD effect did not last very long, necessitating at least two intakes per day. Another problem with this kind of new therapy was that trialists and later on MDs had to learn the hard way just how difficult dosing was: one had to start with a very low dosage and a very low dose increase (low and slow), based upon individual tolerability and, in the case of PD, efficacy. Very frequently patients on bromocriptine had to experience nausea and emesis, dizziness, sedation, and, more frequently than with any other DA agonist, syncope or orthostatic hypotension (linked probably also to the alpha1-adrenergic agonist effects unique for a DA agonist).

The effects of bromocriptine in female hyperprolactinemia were reviewed by Weil (1986). An analysis of 1410 pregnancies achieved with this drug’s help did not show any increase of miscarriages or malformations when compared with normal pregnancies (German information for specialists of November 2016). There was also no increase in postnatal abnormalities in the children—indeed Sandoz once could invite several hundreds of these “bromocriptine-children” with their happy parents for some festivities. When used in PD, however, bromocriptine was less successful as monotherapy, probably as a result of its D1 receptor antagonism (Jackson et al. 1988). Even in combination with levodopa it is being used less frequently than pramipexole or ropinirole, possibly due to its poor tolerability. This includes cases of retroperitoneal or other fibrosis presumably caused by the drug and its active metabolite acting together as mixed 5-HT2B agonist/antagonist (Elenkova et al. 2012). In one case of a macroprolactinoma, retroperitoneal fibrosis associated with long-term bromocriptine therapy did disappear completely once bromocriptine had been replaced by the potent and unsurmountable 5-HT2B antagonist lisuride (Herzog et al. 1989).

Meanwhile, Lisuride hydrogen maleate (Lysenyl SPOFA, Cuvalit, Eunal, Dopergin, Dipergon, Revanil, Arolac, etc.) had been synthesized by Zikán and Semonský (1960) from the Research Institute for Pharmacy and Biochemistry at Prague (later SPOFA) as one of the series of 5-HT antagonists to be used, in a way similar to Sandoz’s methysergide, for the prophylaxis of migraine attacks. It is of interest that one of his pharmacological tests (i.e., the isolated rat fundus stimulated by 5-HT according to Vane) nowadays is known to be specific for 5-HT2B receptors, and lisuride is indeed the most potent antagonist of these receptors (Jähnichen et al. 2006; Hofmann et al. 2006).

When Miroslav Semonský at the Prague Institute of Pharmacy and Biochemistry synthetized lisuride (called Lysenyl in those early days), he very successfully went against three dogmata established by the former Sandoz AG at Basle which presented itself as the expert company when it came to ergot derivatives:

  1. (a)

    that ergot derivatives, to become successful drugs, need to have a peptide side chain;

  2. (b)

    that only derivatives of 8-beta ergot (in those days called Lysergic acid) are effective; and

  3. (c)

    that central effects of ergot derivatives are expression of some toxicity—a belief strongly reinforced by Albert Hofmann’s fortuitous discovery of d-LSD in April 1943 (LSD—Mein Sorgenkind, ed. Klett-Cotta, Stuttgart 1969).

And here we have lisuride—most often used as the hydrogen maleate salt, also for better water solubility which after being registered for migraine prophylaxis as Cuvalit (effective dosage only 3 × 0.025 mg/day) has been approved mostly for its prolactin-lowering and anti-PD effects, i.e., as an 8-alpha-ergoline (a derivative of iso-lysergic acid) drug without any peptide side chain and with prominent central effects. Lisuride has become a very useful drug and, indeed, gold standard for future developments of its derivatives and similar compounds such as pergolide and cabergoline. It also very much differs in its pharmacological effects from the older ergot derivatives: amongst the typical ergot effects as outlined in the Handbook review about ergot (B. Schild and O. Berde, Ergot Alkaloids and Related Compounds, Handbook of Experimental Pharmacology 49, Springer, Berlin, 1978), one finds (1) arterial vasoconstriction, (2) enhanced venous tone, (3) noradrenaline antagonism (resulting, in the case of adrenaline in the famous “Adrenalin-Umkehr” (i.e., reversal of adrenaline-induced vasoconstriction), (4) serotonin (5HT) antagonism and (5) uterotonic effect (caused by a 5-HT agonist effect). As lisuride is completely devoid of (1), (2) and (5), it is somehow unfair to simply call it another ergot derivative, especially as drug authorities and textbooks like to simplify things by “class labels” [which initially had also been applied to all ergot DA agonists when it was observed that pergolide and cabergoline as 5-HT2B agonists could cause cardiac valvulopathy whilst with the 5-HT2B antagonist lisuride, was devoid of any valvulopathy cases in a very large post-marketing data-base of lisuride (Hofmann et al. 2006)]. Lisuride could even be called an “ergot antagonist” because its therapeutic potential consists of highly specific central effects (i.e., its dopaminergic activity) and specific interactions with several 5-HT receptor subtypes, most often opposite to natural ergot derivatives. Lisuride is, indeed, the most potent 5-HT1A agonist known and also very potent 5-HT2A and 2B antagonist, with effective concentrations in the 10−8–10−10 M range, i.e., similar or above its affinities for the DA2 receptor. Lisuride also has high affinities to other G-protein-coupled monoamine receptors and for this reason nowadays is considered to be a multifunctional drug (or, as John Kebabian many years ago had told one of us (RH): “your lisuride is a dirty drug”). It must be noted, however, that high receptor affinity does not always mean high pharmacological activity (as an agonist or antagonist) and, in the case of lisuride, Gessa (1988) has even coined the concept of a high affinity-low activity drug, especially in the case of intact neurotransmission without overactivity or depletion. Furthermore, there are species differences: just as one example, lisuride causes vigourous male-type sexual mounting behaviour in juvenile female rats, associated with ptosis, piloerection and vocalisations similar to behaviour caused by a combination of the inhibitor of 5-HT synthesis pCPA with apomorphine which is independent of the hormonal situation—whether adult male, adult female, juvenile, intact or castrated (Horowski 1983), but there are no human correlates whatsoever of lisuride being an aphrodisiac drug. There are also biochemical similarities of lisuride with d-LSD in rodents (Pieri et al. 1978), but no d-LSD-like hallucinations in healthy people although both drugs stimulate the same central 5-HT2A receptors, but apparently with different post-receptorial pathways (González-Maeso et al. 2007). Here, however, we will deal only with lisuride as a DA agonist (Horowski and Wachtel 1976; Gräf et al. 1976) as these effects of lisuride have been reviewed several times quite extensively (so by Horowski and McDonald 1983; by G Gopinathan, R Horowski and IH Suchy in the Handbook of Experimental Pharmacology 88, Drugs for the treatment of Parkinson’s Disease, DB Calne ed., Springer, Berlin, pp 471–513, 1989; by R Horowski and JA Obeso, pp 219–248 in Therapy of Parkinson’s Disease, WC Koller and G Paulson eds., Marcel Dekker New York, 2nd ed. 1995 (mostly about parenteral application) and in a full monography: Lisuride and other Dopamine Agonists, DB Calne, R Horowski, R McDonald and W Wuttke, Raven Press New York, 1983). There is also a special issue of the J Neural Transm Suppl 27, JA Obeso, R Horowski, DC Marsden: Continuous Dopaminergic Stimulation, Springer Wien, New York, 1988, again about the parenteral use of lisuride. The basis for the long-lasting prolactin-lowering effect of lisuride (Liuzzi et al. 1978) is caused by lisuride stimulating D2 receptors on the prolactin cells [where lisuride, due to its high receptor affinity, accumulates (M. Hümpel, pers. comm.)]. Lisuride was also the first prolactin-lowering drug where a shrinking of anterior pituitary macroprolactinomas has been observed (Chiodini et al. 1980; Liuzzi et al. 1985). After the approval of lisuride as a prolactin-lowering drug by the German regulatory authority in 1979, within the first years more than 1000 well-documented pregnancies from women with prolactin-associated infertility could be obtained in Germany alone, again in their outcome with no differences to other spontaneous pregnancies. It has been estimated by experts from Novartis that up to now about 100,000 pregnancies have been achieved since Kjell Fuxe first has described the prolactin-lowering effects of dopaminergic ergot derivatives. The anti-PD effects of lisuride depend mostly on the stimulation of D2 receptors (Schachter et al. 1980). All therapeutic applications of lisuride so far (after its use for the prevention of migraine where lisuride has been withdrawn from the market for too low sales) are based upon the dopaminergic activity of this drug. Oral use of lisuride is complicated by its high and is worse, variable first pass effect caused by the different p450 enzyme status. Therefore, it has to be up-titrated very slowly, with increases by just 0.1 mg per week, until an effective dosage has been reached and also until tolerance has developed towards so-called early adverse events such as nausea, emesis, dizziness and orthostatic hypotension. As other DA agonists such as pramipexole do not have this problem, lisuride, which once was marketed in nearly all European and a number of Latin-American countries, after the take-over of Schering AG/Berlin by Bayer, was progressively withdrawn from all markets (never due to regulatory or safety reasons but always due to poor sales). Obviously, the pharmacokinetic problems of oral lisuride could be overcome by parenteral and transdermal application (Stocchi et al. 2002; Woitalla et al. 2004), and in this way, lisuride becomes an excellent drug to achieve continuous dopaminergic stimulation (CDS), which is considered to be the best therapeutic option for early and longterm treatment of PD (for review see Olanow et al. 2006). Such a parenteral use of lisuride had been proposed and used successfully in PD patients by the Vienna group (Seemann et al. 1984, Walther Birkmayer and Peter Riederer, personal communication 1979). The great therapeutic potential of lisuride as an antifibrotic 5-HT2B antagonist as predicted by animal models, but also the use of this fascinating drug as a DA agonist in new application forms (patch, patch pump, portable mini-pumps for s.c. infusion or others) is currently being studied.

There are two other concepts developed by A. Carlsson on the mechanism of action of DA agonists: the concept of a presynaptic agonism to reduce DA function in schizophrenics (Lahti et al. 1998). Next he investigated the DA partial agonists for further therapeutic use (Burstein et al. 2011). These studies were based on earlier results suggesting that DA partial agonists may have therapeutic potential in management of side-effects of antipsychotic treatments (Svensson et al. 1993). In line with these observations it has been demonstrated that the lisuride derivative terguride, a mixed dopamine (DA) agonist/antagonist on normosensitive receptors but showing DA agonistic properties at supersensitive DA receptors, is effective in primate models of PD. Such a compound could offer an alternative to the treatment of Parkinson’s disease with indirect or direct DA agonists (Lange et al. 1992). Interestingly, the lisuride analogue brominated in position 2 acts as DA antagonist with potentially atypical neuroleptic properties (Löschmann et al. 1992).

Taken together the pioneering work of A. Carlsson led to rationale therapy of patients suffering from PD and a broad spectrum of potential pharmacological interventions in a variety of central nervous system disorders. A number of them remain to be explored and may eventually become a valuable therapeutic option for patients suffering from a variety of diseases.