Synthesis of a Novel Type of 2,3′‐BIMs via Platinum‐Catalysed Reaction of Indolylallenes with Indoles

Abstract Optimisation, scope and mechanism of the platinum‐catalysed addition of indoles to indolylallenes is reported here to give 2,3′‐BIMs with a novel core structure very relevant for pharmaceutical industry. The reaction is modulated by the electronic properties of the substituents on both indoles, with the 2,3′‐BIMs favoured when electron donating groups are present. Although simple at first, a complex mechanism has been uncovered that explains the different behaviour of these systems with platinum when compared with other metals (e.g. gold). Detailed labelling studies have shown Pt‐catalysed 6‐endo‐trig cyclisation of the indollylallene as the first step of the reaction and the involvement of two cyclic vinyl‐platinum intermediates in equilibrium through a platinum carbene, as the key intermediates of the catalytic cycle towards the second nucleophilic attack and formation of the BIMs.


General experiment details
All reagents were purchased from commercial sources and used without further purification, unless stated otherwise. All reactions were carried out under nitrogen atmosphere and in the absence of moisture, unless stated otherwise. Reactions using microwave irradiation were carried out in Biotage Initiator+ Microwave system. Reactions were monitored using Thin Layer Chromatography (TLC) using 0.2 mm thick silica gel plates 60F-254 (5735 Merck) with a mobile phase of hexane and ethyl acetate, with visualization by illumination by uv light λ = 254 nm or staining with either potassium permanganate or phosphomolybdic acid solution. 1 H NMR and 13 C NMR spectra were recorded on a Bruker (500 MHz) spectrometer with CDCl3 solvent. Chemical shifts (δ) are given in parts per million (ppm) and coupling constants values (J) are given in Hertz (Hz), and are approximated to the nearest 0.1 Hz. 13 C NMR was recorded using broad-band proton decoupling. Abbreviations used in NMR analyses are as follows: s = singlet, d = doublet, t = triplet, q = quartet, p = pentet, dd = doublet of doublets, dt = doublet of triplets, td = triplet of doublets, ddd = doublet of doublet of doublets, dq = doublet of quartets, qd = quartet of doublets and m = multiplet. HRMS were performed by EPSRC National Mass Spectrometry Service Centre, Swansea.

Optimisation
Screening of solvent, time, temperature, concentration and equivalents of platinum, indole and methanol were carried out for the reaction of 3-methyl-N-(2,3-butadienyl)indole 1a with external indole 2a to optimise the conditions for selective formation of BIM 3aa. These reactions were carried out at the Novartis laboratories using LCMS ELSD (Evaporative Light Scattering Detector) traces for detecting the different components of the reactions.
Reported reactions by Muñoz et al. 1 were carried out at 70 o C using thermal heating in an oil bath. However, it was established that using microwave irradiation decreased the reaction time dramatically from 20 hours to 1-2 hours as well as increasing both the selectivity and yield of product 3aa. Microwave irradiation was therefore used for all further screening and extensive screening of solvent and temperature was carried out with THF, 1,4-Dioxane, CPME (cyclopentylmethyl ether) and 2-methyl THF (   Variations in equivalents of platinum, indole and methanol were then explored. We found that although results were slightly better with PtCl2 10 mol %, similar reactivity was achieved when 5 mol % was used, so this was chosen as the optimized conditions. Table 2.2 shows the results when different amounts of methanol were used, with the best results achieved with 3 or more equivalents as already reported (entries 3-5). The equivalent of indole to allene was also investigated and Table 2.3 shows that using 3 or 4 equivalents gives the best conversion to product 3aa.  After this screening we chose the optimised conditions as: microwave irradiation at 130 o C for 1 hour with 5 mol % PtCl2, 3 eqs indole, 3 eqs of methanol and 1,4-dioxane (0.2 M). These newly optimised conditions were able to give an isolated yield of 61 % for compound 3aa. It is important to state that the reaction requires dry 1,4-dioxane as the solvent as reactions carried out with wet 1,4-dioxane under optimised conditions resulted mainly in the formation of the cyclised products 4a/4a' ( Table 2.4).

General procedure for platinum-catalysed reaction of indolyl allene with external nucleophile under optimised conditions
PtCl2 (5 mol %) and the appropriate nucleophile (3 eqs) were added to a microwave vial, capped and flashed with N2 atmosphere. The solids were dissolved in dry 1,4-dioxane and the appropriate indolylallene (1 eq) dissolved in dry 1,4-dioxane (0.2 M) was added. Dry methanol (3 eqs) was added and the vial was heated under microwave irradiation at 130 o C for 1 hour, or the appropriated time indicated for each particular compound. The resulting reaction mixture was filtered through Celite and washed with DCM and compounds purified via column chromatography in silica gel with Pet/EtOAc.

Dimers
In reactions with external indoles (entries 8, 10, 11 and 12, Table 1) and in the reaction with the ethyl derivative in C3 of the indolylallene (entry 2, Table 2), small amounts (3-10 %) of a dimer were isolated (see tables 1 and 2 in main manuscript). The structure of the dimers was confirmed by HRMS and tentatively assigned by NMR analysis ( 1 H, 13
To a pre-washed round bottomed flask was added 1-methylindole (1220 mg, 9.3 mmol), 2.5 mL D2O was added, the reaction mixture was heated to

Reaction profile of 1a with d-2i monitored by 1 H NMR
Platinum-catalysed reaction was carried out under optimal conditions and microwave irradiation with the previously isolated 6-endo cycle 4a and external N-methyl indole d-2i with deuterium incorporated into position 3 of the indole (Figure 5.3.1) using dimethylsulfone as internal reference. Samples of the reaction mixture were taken every 10 minutes over a 90minute period and the progress of the reaction was analysed by 1 H NMR of the crude using the signals corresponding to the protons in position a of the three compounds: 4 ppm (CH2, 4a), 4.5 ppm (CH2, 4a') and 4.18 ppm (CH diastereotopic, 3ai) to measure the integrals/concentrations of products 4a, 4a' and 3ai over time using the signal of the dimethylsulfone as reference. The plot the concentration of the products over time (Figure 5.3.2) shows that the nonconjugated cycle 4a isomerizes to the conjugated cycle 4a' within the first 20 minutes. Also noted is that as soon as isomerisation to cycle 4a' begins, so does the formation of compound 3ai, although at a slower rate. At 20 minutes, concentration of cycle 4a' starts to decrease as the concentration of 3ai continues to increase and once both of the cycles are consumed the reaction is complete. The reaction was carried out with deuterated N-methyl indole d-2i, with 90 % deuterium incorporation at the beginning of the reaction. After 5 minutes we observed a loss of deuterium in position 3. Overall we observe a deuterium loss in the external indole of 41 % from position 3 and analysis of the NMRs shows that deuterium is incorporated into cycle 4a' at position d within the first 10 minutes. Analysis also shows that deuterium is incorporated into position d of compound 3ai after 10 minutes with around 31 % deuterium incorporated in the final compound.

Deuteration experiments
Reactions were carried out using deuterated and undeuterated starting materials 1a, d2-1a, d-1a, 2i, d-2i or a combination of them in the absence and presence of methanol, either CH3OH or CD3OD (Table 6.1). The ration of the three products and the deuterium incorporation in the different positions observed for all the compounds are indicated for each reaction. The deuterium incorporation was analysed by measuring the integrals of the different signals of all the products in the 1 H NMR of the crude and by analysis of the 1 H NMR and the HSQC of the purified compounds when possible.
Representative analysis of a pure sample of 2,3'BIM dn-3ai obtained in the reaction in entry 1 in Table 6.1 is shown below: Analysis of the 1 H NMR of the purified dn-3ai in figure 6.1a shows deuterium incorporation in positions b to d, this was established by comparing the spectrum of the deuterated product with the non-deuterated example, analysing the disappearance of signals or change in multiplicity on those positions and measuring the integrals in relationship to the signal of the proton at 6.25 ppm corresponding to one proton on the C2 position on the N-methyl indole. Position a is unaffected in the reaction, whereas positions b to d have significantly lower integrals which supports deuterium incorporation. Interestingly, both diastereotopic positions of methylenes b and c showed some degree of deuterium incorporation. These are also supported by analysis of the HSQC (Figure 6.1b) where we observe a mixture of CH2 (blue spots) and CHD (red spots, shifted to the top right of the blue signals due to the isotope effect), the intensity of these two signals are in proportion to the calculated % D incorporation.

6.2.a. Synthesis of 13 C-1a
13 C-labelled 3-methyl-N-(2,3-butadienyl)indole 13 C-1a, with the 13 C in the terminal carbon of the allene, was synthesized by standard Crabbé homologation using 13 Cparaformaldehyde. 13   The reaction was monitored by 1 H NMR and 13 C NMR over a 90-minute period, with samples taken every 10 minutes and the progress of the reaction was analysed by 1 H NMR of the crude using the signals corresponding to the protons in position a of the three compounds: 4 ppm (CH2, 4a), 4.5 ppm (CH2, 4a') and 4.18 ppm (CH diastereotopic, 3ai) to measure the integrals/concentrations of products 4a, 4a' and 3ai over time using the signal of the dimethylsulfone as reference. Analysis of the 1 H NMR spectra of the crude of the samples showed very fast reaction of the allenyl indole 1a to form cycle 4a and isomerization of cycle 4a to 4a' as seen before. However, in this case full conversion of 64  65  66  67  68  69  70  71  72  73  74  75  76  77  78  79  80  81  82  83  84  85  86  87  88  89  90  91  92 f1 (

Reaction of 3ai in the presence of PtCl2 and CD3OD
The isolated non-labeled 2,3'-BIM 3ai was reacted under platinum conditions in the presence of deuterated methanol to determine if deuterium incorporation can occur further at position d in an out-of-cycle process (Scheme 6.3). The 1 H NMR after 1 hour shows 42 % deuterium incorporation at position d only (Figure 6.3