Combinatorial Chemistry and SPOS:
Solid phase synthesis was soon adopted in nucleic acid chemistry, where the chemistry revolves around repeating units of nucleotides linked through phosphodiester bonds. More recent developments in solid phase synthesis are the application of the methodology to combinatorial chemistry and organic syntheses of small molecules. We would discuss these recent developments here.
In combinatotial chemistry, several closely related molecules are synthesized in one pot, thus generating molecular libraries of organic molecules. A library could contain any class of organic molecules. This concept was first exploited in creation of peptide libraries. Bulk of the literature on Combinatorial Chemistry centers around peptides. Solid Phase Organic Synthesis (SPOS)' refers particularly to creation of small organic molecule Libraries, synthesized using solid phase methodology. These procedures permit synthesis of well-defined molecular libraries very quickly and in very pure form. Let us have a very brief look at these two developments. The general principles would be developed around peptide chemistry.
Tea Bag Technique: A new era of combinatorial chemistry was born in 1985, when Houghton published his Tea Bag Technique that eventually allowed synthesis of several peptides in pure form in remarkably short time. This method was referred to as the "tea-bag" method because in the first reported procedure, the peptides were enclosed in mesh bags and dipped quickly into liquid solutions containing bacteria or virus cells. The peptides which proved most capable of binding to cells were retained. The process was repeated until a single peptide strain was selected. This allowed "[the capture of] information in a day that you couldn't get in a hundred years before" according to Houghten. In 1985,
Houghten, Richard A. (August 1, 1985). "General Method for the Rapid Solid-Phase Synthesis of large Numbersof Peptides: Specificity of Antigen--Antibody Interaction at the Level of Individual Amino Acids".Proceedings of the National Academy of Sciences
Houghten used small polypropylene mesh bags which were big enough to make about 500 μmoles of peptides. They took advantage of the fact that several operations in peptide synthesis were common procedures. Each bag was well labelled with tags that could withstand through the chemical processes. Each bag was intended for the synthesis of one peptide sequence. All bags were combined for routine wash cycles and deprotection steps. The bags were then
segregated into groups depending on the amino acid that went into the next step. This procedure saved chemicals and time, enabling the synthesis of a large number of pure peptides in remarkably short time. Thus 260 peptides, about 10-20 mg each, were made in 4 weeks.
In 1988 and 1991 Furka reported a method that changed the way we looked at synthesis of drugs and screening of drugs To understand this chemistry
Furka, Á., Sebestyén, F., Asgedom, M., Dibó, G., Proceedings of the 10th International Symposium of Medicinal Chemistry, Budapest, Hungary, 1988. p. 288. Furka, Á., Sebestyén, F., Asgedom, M., Dibó, G., Int. J. Peptide Protein Res., 37 (1991) 487.
better, let us have a close look at one ‘bead’. A single bead has more than one ‘reactive site’ where the syntheses takes place. Each bead behaves independently during synthetic operations holding one pure compound. This is refered to as ‘one-bead-one-compound’ concept. Let us look at one reactive site attached to one bead in the flask. For a classical synthetic operation, all beads in one flask would have same molecular units attached during a coupling step. Now let us mix two such reaction flasks into one flask. Each bead would maintain its identity even when beads from different flasks are mixed. Once mixed, we cannot tell one from the other (unless we use special techniques).
Now let us have a close look at Mix and Split Synthesis. Three batches (of beads) are first linked to three different amino acids A, B and C in independent reaction pots. After the wash sequence, the three batches are mixed thoroughly, split into three batches again and coupled with three amino acids as before (Mix – Split – Couple 1). Now repeat the mix – split - couple operation as shown in Fig 5.1.
Fig 5.1 Mix / Split / couple protocol for synthesis
Thus, in the first mix – split – couple process we have a mixture of 9 molecules in three pots. Now look at the economy even at the dimer stage. Classical parallel syntheses would require 9 sets of couple – wash experiments to get 9 dimeric compounds. By this new procedure, we do only two couple – wash steps to get 9 compounds. This amounts to a huge saving in time, apparatus, manpower and chemicals, all of which translate into a huge amount of money.
The library grows very fast indeed. Simple calculations suggest that if only small peptide libraries are desired using all the 20 proteinogenic amino acids, the number grows exponentially as shown below (Fig 5.2).
Fig 5.2: Number of dipeptides synthesised using any of the 20 amino acids.
The utility of combinatorial chemistry to pharma industry could be gauged from the following data Fig 5.3.
Fig 5.3: Power of combinatorial synthesis.
Screening: Since the compounds are not pure individual compounds, how does one screen the mixture to arrive at one active compound? There are several methods through planned synthesis of libraries and screening. Let us look at just one process called Iterative Screening procedure (Houghton (1991)). Let us consider a batch of three tripeptide libraries synthesized above. First screening of the three tripeptide pots tells that pot 2 has an active compound. Since ‘Y’ is the last amino acid that was coupled in this batch, the corresponding dipeptide pots are coupled with ‘Y’alone and screened. This time it says pot one is active. Since this pot has only three peptides, the samples are synthesized independently and tested to identify the active molecule (Fig 5.4). Thus, billons of compounds could be synthesized and screened in less than a couple of months.
Fig 5.4: Iterative screening by Houghton (1991)
The screening procedure discussed above tested the beads directly. This is indeed a fast process. In microbiological screening, the bead that shows inhibition to growth of the organism could be hand picked (with tweezers of course) and the structure of the peptide that is attached to the bead identified using Single Bead FT-IR Microspectroscopy, NMR (Magic Angle Spinning NMR) and MS spectroscopy. Alternately, the bead itself could serve as a label. Different beads could be used for different families. Thus, by identifying the bead, one can tell the family. The spacer and linker could be labeled. The label could be an unnatural amino acid, a special chain, a terpene that would give special colour tests, a compound that carries a radio label, phosphorescent tags, fluorescent tags, UV tags etc., that could serve as marker for specific family (families). A cleaver combination of labels could narrow down the search.
This structural analysis assumes that only one side of the bead is exposed to the biochemical interactions due to microorganisms. This is generally true. The problem with this technique is that a negative test does not mean that the library is rejected. A negative result could very well be due the insufficient exposure of the target molecules to the molecules on the bead.
An ideal procedure would be to delink the library and look for bioactivity of the library in solution phase. The same Iterative screening procedure could be used on solution mixture as well. Houghton’s group screened for an active peptide of the general formula Ac-XXXXXX-NH2, where X = any of the 18 or 19 proteinogenic amino acids. The amino acids cystein and tryptophene were not used in the first and second positions. The assay was carried out using a competitive Enzyme Linked Immunosorbant Assay (ELISA) to identify the mixture that caused the greatest inhibition of the antibody.
In the Synthetic Peptide Combinatorial Library (SPCL), two positions were initially defined (O1 and O2), while the remaiming four positions were equimolar mixture of 18 amino acids. Thus 324 SPCLs (AcO1O2XXXX-NH2) were analysed and the library having D at position 1 and V at position 2 showed best results. The next SPCLs had the 3rd position defined by the 20 amino acids, while the remaining three positions were XXX, consisting of equimolar mixture of 19 amino acids. This gave 193 (6,859) peptides and screening revealed the position 3 was for P. similarly, libraries Ac-DVPOXX-NH2 , Ac- DVPDOX-NH2 and Ac-DVPDYX-NH2 were screened. The last member was identified by synthesizing all 20 hexapeptides to give Ac-DVPDYA-NH2 as the most active peptide. A total of 34,012,224 peptides were thus analysed and the sequence identified. The merit of this procedure is that there are no assumptions in the search process.
Another procedure described by Brondelle (1996) is called Positional Deconvolution Process. Testing for anti-staphylococal activity was carried out on the following permethylated hexapeptides. The sequences and the results are given in the Table below (Table 5.5).
There are limitations and merits in each of these procedures. Several other screening procedures are known. The most important point to note is that the time taken for discovery of a drug candidate is brought down drastically. Once the candidate is identified, the same procedure allows the synthesis of several analogues. Screening them by special techniques permits discovery of new drugs with most desirable characteristics.