lasted for many years. Still these hallmark results form the foundation of chemotherapy and led to the development of other alkylating agents like chlorambucil, melphalan, and cyclophosphamide (Figure 1.2), which are better tolerated and are still used today in clinical practice. It is noteworthy to correct a historical mistake that is frequently made. The bombing of a ship in Bari during World War II, which led to exposure of the crew to mustard gas, is often cited as the discovery of mustard's antitumor activity and the discovery of chemotherapy. This is not correct. Despite the fact that severe leucopenia was also observed in affected soldiers, the German air raid on the ships in the harbor of Bari took place on 2 December 1943, more than a year after patient J.D. had been treated. The development of chemotherapy is a fascinating topic, which has been reviewed in appropriate detail elsewhere [17].
Figure 1.2 S‐Lost, N‐Lost, modern agents.
1.3 Pregabalin
The discovery of pregabalin by Richard Silverman [18] and coworkers is a great example of successful identification of a small‐molecule drug in academia. γ‐Aminobutyric acid (GABA) was recognized early on as an important inhibitory neurotransmitter in the brain (Figure 1.3) [19]. The observation that GABA levels and L‐glutamic acid decarboxylase (GAD) activity is decreased in a number of pathologies like epilepsy, Alzheimer's, and Parkinson's disease has sparked the search for drugs to increase GABA levels in the brain. Pursued strategies include development of GABA receptor agonists, GABA uptake inhibitors, and inhibitors of 4‐aminobutyrate‐oxo‐glutarate aminotransferase. The latter enzyme is the key catabolic enzyme of GABA. Inhibitory effects of hydroxylamine on γ‐aminobutyric acid aminotransferase (GABA‐AT) were described already in 1961 [20]. In 1966, inhibition of (GABA‐AT) by aminooxyacetic acid was disclosed [21].
It was also demonstrated that inhibitors available at the time demonstrated insufficient selectivity [22]; consequently this approach was rendered as likely to be unsuitable to target epilepsy in humans. Shortly after starting his own laboratory at Northwestern University in Illinois in 1976, Silverman got interested in the biology of GABA‐AT and set out to develop chemical inhibitors. He published his first manuscript on the subject as early as 1980 [23]. While his first efforts relied on optimization of irreversible inhibitors, he was not able to overcome the intrinsic non‐specificity of these compounds. Specifically, inhibition of L‐glutamic acid decarboxylase (GAD) turned out to be an issue. GAD catalyzes the conversion of L‐glutamate, an excitatory neurotransmitter to the inhibitory neurotransmitter GABA. Inhibition of GAD would consequently lead to a decrease in GABA concentration and thus be highly undesirable.
In 1988 a visiting postdoc, Riszard Andruskiewicz from Gdansk University, joined Silverman's laboratory and was asked to work on synthesis and characterization of 3‐substituted GABA and glutamate analogues. He synthesized a set of 14 3‐alkyl‐GABA derivatives (Figure 1.4), 4‐methyl GABA, and the two enantiomers, as well as seven glutamate derivatives. Most interestingly and also somewhat surprisingly, all of the GABA analogues were found to be activators of L‐glutamic acid decarboxylase [24].
Figure 1.3 GABA biology.
Figure 1.4 3‐Me‐GABA analogues synthesized by Andruskiewicz and Silverman.
At that point (1989), they filed an invention disclosure and engaged in discussions with potential industrial partners, which led to start of collaborations with Upjohn Pharmaceuticals and Parke‐Davis Pharmaceuticals. The most potent compound, (R)‐3‐methyl‐GABA, did not display convincing anticonvulsant activity. Upjohn, concentrating on profiling the “best” compound, ended the cooperation at that point, while Parke‐Davis scientists tested all derivatives and found that the isobutyl derivative resulted in very favorable pharmacological effects. This was somewhat surprising, as the activation of GAD was significantly weaker for this compound compared with the corresponding methyl derivative (R/S)‐methyl‐GABA (239 % activity of GAD at a concentration of 2.5 mM versus 143 % activation for the racemic isobutyl analogue) [24]. However, after synthesizing the two isobutyl enantiomers, they could confirm that (S)‐3‐isobutyl‐GABA, later named pregabalin (Lyrica™), displayed one of the most pronounced anticonvulsant activities they ever tested. Several years later, Parke‐Davis scientists demonstrated that pregabalin binds to Ca2+‐channels, subsequently inducing calcium flux into the neuron. In turn this resulted in inhibition of glutamate and substance P secretion from excitatory neurons. So, in fact, the mechanism underlying the observed pharmacological effect of pregabalin, which was thought to be mediated by inhibition of GAD, was completely different. Inhibition of glutamate secretion does result in a similar pharmacological effect. Also the enhanced potency compared with other related derivatives could be explained by pregabalin being a substrate for the System L transporter, enabling active uptake into the brain [25]. Other compounds, like GABA itself, are not substrates of this transporter. Thus, their capability of crossing the blood–brain barrier is very limited.
Interestingly, in principle it only took the synthesis of 16 compounds to initiate the development of a successful drug candidate. Certainly many more compounds were produced and characterized in the Silverman laboratory, and still, the development of the actual drug required another 15 years until it was finally approved by the FDA in December 2004. But this drug development represents one of the rare cases where the final molecule was already obtained early on in the project. The originally assumed optimization rationale turned out to be not the correct one in various aspects, but by careful pharmacological examination, pregabalin was identified. This underlines the necessity to remain open to unexpected findings and keep the flexibility of adapting optimization goals and target values, or even the optimization strategy as a whole.
As a part of the deal with Silverman and the university, Pfizer (which had subsequently acquired both Park Davis and Upjohn), agreed to pay 4.5 % of global sales to the university, and Richard Silverman, who split his share with his coworker Andruszkiewicz, would receive 1.5 %. As Lyrica turned into a real blockbuster molecule, the university received an estimated US$ 1.4 billion in royalties.
On the topic of academic drug discovery, Silverman wrote in 2016: “Academic scientists are not constrained by the requirement of making products to remain viable; therefore, shortcuts are not necessary, and tangential observations can be explored, which may lead to new discoveries. Because of this, academic invention needs to be encouraged in all areas of pursuit to allow new products to become available to society; industry should assist in financing the development of these products.” [18]
1.4 Natural Product‐Derived Drug Discovery
Another important contribution of academia to drug discovery is providing specific expert knowledge on particular research areas and techniques. This knowledge, acquired within the academic group of a professor throughout his complete academic career, may represent the long‐sought solution to a specific problem that hampers progression of a compound to the market or prevents it from moving into clinical trials. This can be of particular value in the field of natural product research, as structural complexity is tremendous
