The inverse-electron demand Diels–Alder reaction of tetrazines with cyclic enol ethers

Inverse electron-demand Diels–Alder additions (iEDDA) between 1,2,4,5-tetrazines and suitable unsaturated dienophiles such as olefins, alkynes, or enol ethers provide facile access to pyridazines. Herein the use of cyclic enol ether derivatives for preparing pyridazines bearing 2-hydroxyethyl, 3-hyproxypropyl, and 3-oxopropyl substituents at the 4-position is disclosed and second order rate constants for the reactions with 2,3-dihydrofuran, 3,4-dihydro-2H-pyran, and 2-methoxy-3,4-dihydro-2H-pyran are presented.


Introduction
Since the first report of the synthesis of pyridazines from tetrazines and alkenes, allenes, or alkynes by Carboni and Lindsey in 1959 [1], this reaction received ever increasing attention first because of its synthetic potential in heterocyclic chemistry [2] and later in chemical biology applications because particular tetrazine/olefin combinations provide exceptionally fast and selective ligation reactions at very low concentrations [3].Additionally, applications in material chemistry evolved [4][5][6].The mechanistic understanding of the inverse-electron demand Diels-Alder cycloaddition (iEDDA) reactions, the broader term for the reaction of a tetrazine with an olefin giving a pyridazine derivative, is well developed and nicely summarized in a recent review article [7].
A special case of dienophiles are enol ethers.Enol ethers undergo iEDDA and subsequently the primary formed dihydropyridazine product aromatizes upon fast or even spontaneous elimination of the corresponding alcohol [8][9][10].This reactivity has been exploited in natural product syntheses [11,12], but in particular in chemical biology in so-called click to release reactions [3].A click to release reaction is a bioorthogonal bond-cleavage reaction, in this case caused by the iEDDA of tetrazine with an enol ether, releasing the corresponding alcohol.In that way, e.g., drug release can be triggered [13][14][15] or caged fluorophores can be unmasked and used as bioorthogonal fluorogenic probes [16].
Dedicated to the memory of Professor Fritz Sauter. 1 3 Herein we wish to report the use of cyclic enol ethers as the dienophile in iEDDA reactions with tetrazines for preparing pyridazines with 2-hydroxyethyl or 3-hyproxypropyl substituents in 4-position of the pyridazine ring as an alternative for hitherto used alkynes (Scheme 1).Earlier work of Roffey and Verge [8], Sauer et al. [9], and Boger et al. [17] presented some early examples of using cyclic enol ethers.Taking into account, that many reliable and high yielding synthetic strategies for preparing many differently substituted tetrazines became available in the last years [18][19][20], access to many new pyridazine derivatives is facilitated by the herein disclosed methodology.

Results and discussion
As a starting point, the reaction of 3,6-di(pyridin-2-yl)-1,2,4,5-tetrazine (1a) with 2,3-dihydrofuran (2a) was investigated under different reaction conditions.Firstly, the impact of excess of olefin 2a on the time needed for full conversion of tetrazine 1a was studied.A fast reaction of an equimolar amount of the starting materials at room temperature is desirable for qualifying the reaction as a click reaction.However, at 23 °C in CH 2 Cl 2 as the solvent the reaction is rather slow giving full conversion of 1a in about 12 h (Table 1, entry 1).Upon increasing the excess of 2a, the reaction becomes faster and upon using 10 equiv. of 2a, full conversion of 1a can be obtained after only 10 min (Table 1, entries 2-4).Increasing the temperature to 40 °C and using only a slight excess of 2a (1.2 equiv.)gave full conversion of 1a after 70 min (Table 1, entry 5).A further rise of the temperature to 55 °C (the boiling point of 2,3-dihydrofuran is 54-55 °C [21]) led to reaction times of 15-25 min depending on the solvent used.In less polar toluene, the reaction is distinctly slower than in dioxane or THF at the same temperature (Table 1, entries 6-8), which is in stark contrast to the iEDDA of styrenes with 1a.In these cases, the reactions in toluene were approx.1.5 to 2 times faster than in THF [22].Using toluene at 100 °C a similar reaction speed than obtained in THF at 55 °C can be obtained (Table 1, entry 9).
In all cases, a single main product formed, which is in every case accompanied by side products up to a share of about 10% in sum.The main product was readily identified as 4-(1-hydroxyethyl)-3,6-di(2-pyridyl)pyridazine (3a) upon comparison with published data [23].The side products were found to be 3,6-di(pyridin-2-yl)-dihydrotetrazine (3a′) and most probably 4,7-di(pyridin-2-yl)-2,3dihydrofuro [2,3-d]pyridazine (3a″), see Scheme 2. 3a′ and 3a″ might result from the oxidation of initially formed 2,3,3a,7-tetrahydrofuro [2,3-d]pyridazine intermediates by remaining 1a.Such an oxidation of dihydropyridazine derivatives by tetrazines is known [24,25] and would explain the similar amounts of 3a′ and 3a″ observed in the crude product mixture.Actually all attempts to suppress the formation of the side products failed (Table 1).Also running the reaction under inert atmosphere of N 2 gave similar results, so that oxidation by oxygen from air, used to oxidise dihydropyridazines [26], can be ruled out as reason for the formation of 3a″.However, in the present case, aromatization of the pyridazine ring upon ring-opening of the 2,3-dihydrofuro moiety (i.e., elimination of the alcohol) is faster than oxidation and consequently 3a is the by far prevailing product.Monitoring the reaction at early stages using 1 H NMR spectroscopy reveals the intermittent appearance of resonances tentatively assigned to 4,7-di(pyridin-2-yl)-2,3,3a,7a-tetrahydrofuro [2,3-d] pyridazine (ii in Scheme 2) the intermediate created upon extrusion of N 2 from the initially formed tetraazabarrelene derivative (denoted i in Scheme 2).Quantum chemical calculations on related reactions support the assignment, as the formation of the primary product i is found exothermic and the barrier for obtaining the distinctly more exothermic intermediate ii is usually rather small [27][28][29].No other side products or intermediates were observed.Accordingly, the preparation of 3a from 1a and 2 is most economically performed with a slight excess of 2 at elevated temperature in THF (according to Table 1, entry 8).Purification was done by column chromatography and 70% yield were obtained.For comparison, the previously disclosed synthesis of 3a starting from 1a and 3-butyn-1-ol (2 equiv.) was performed by heating a solution in toluene at 110 °C for 75 h.In this case, 3a was obtained in 77% yield after a purification step by column chromatography [23].Accordingly, the use of 2a provides a distinctly faster and resource saving access to 3a.
In a next step, the higher homologue 3,4-dihydro-2H-pyran (2b) was used as dienophile (Scheme 3).Testing revealed a distinctly lower reactivity when compared with 2a.At room temperature, no reasonable reactivity could be obtained and a solution of 1.2 equiv.2b in toluene heated to 100 °C needed 16 h for full consumption of 1a.Using 5 equiv.2b at otherwise same reaction conditions needed 5 h for completion.The product 4-(1-hydroxypropyl)-3,6-di(2pyridyl)pyridazine (3b) was isolated in 70% yield after a column chromatographical purification step necessary for the separation of minor amounts of 3a′.A side product similar to 3a″ was not observed.
Switching to the less electron poor and thus less reactive tetrazine derivative 3,6-diphenyl-1,2,4,5-tetrazine (1b) distinctly slower reactions were observed in all cases.With 1.2 equiv.2a full conversion of 1b is obtained in approx.5 d using toluene as the solvent and a reaction temperature of 100 °C.When using excess of 2a and no further solvent, the reaction time can be shortened to about 8 h.The desired product 4-(1-hydroxyethyl)-3,6-diphenyl-pyridazine (4a) can be isolated upon separation of minor unidentified impurities by column chromatography in 67% yield.Similarly, 4-(1-hydroxypropyl)-3,6-diphenyl-pyridazine (4b) needs excess of 2b for a reasonable short reaction time.After heating the reaction mixture for 48 h at 100 °C and similar workup as in the case of 4a, 4b was isolated in 53% yield.Tetrazine 1b can also be used to prepare 3-(3,6-diphenyl) pyridazin-4-yl)propanal (4c).In this case, the reaction is carried out in neat 2c for 52 h at 110 °C and after workup, 39% of the desired product is obtained.
To set the reactivity of cyclic enol ethers into relation, the second order rate constants of their reactions with 1a and 1c were determined.For electron richer 1b, not featuring intramolecular repulsive N-N interactions, recently identified as the decisive effect for accelerating the cycloaddition step in case of tetrazines 1a and 1c [30], rate constants were not determined because the reactions are very slow under the chosen measurement conditions.Results are shown in Fig. 1.The five-membered cyclic enol ether 2a reacts approx.200 times faster than its 6-membered homologue 2b and about 1000 times faster than 2c.An aspect explaining this are the different ring strains of 2a and 2b as it has been shown, that the reactivity of dienophiles increase with increasing ring strain [31].A similar trend has been found when using highly reactive 1,2,4,5-tetrazine-3,6-dicarboxylate (with 2a: k = 437 mol −1 dm 3 s −1 ; with 2b: k = 1.75 mol −1 dm 3 s −1 in 1,4-dioxane) [9].
The electron poorer tetrazine 1c reacts roughly ten times faster than 1a, however 1a and 1c are investigated in different solvents.When comparing the second-order rate constants in the same solvent, a higher difference was noted as k for the reaction of 1a with 2a in chloroform is only 0.011 ± 0.0005 mol −1 dm 3 s −1 .The solubility of 1c in methanol is too low to permit the rate constant determinations in this solvent.
In comparison to commonly used dienophiles, it is revealed that 2a exhibits a second-order rate constant about an order of magnitude higher than styrene (with 1a k = 0.003 mol −1 dm 3 s −1 in methanol) [32].Cyclopentene gives a slower and norbornene a faster reaction (with 1a k = 0.008 resp.0.15 mol −1 dm 3 s −1 in methanol) than 2a [32].Dienophiles 2b and 2c can be regarded as rather slow reactants in this reaction.However, when compared to terminal alkynes, their second-order rate constants are still about an order of magnitude higher [9].

Experimental
Chemicals were purchased from Sigma-Aldrich, Fisher Scientific, Merck, or Alfa Aesar.All reagents were used without further purification unless otherwise noted.Tetrazines 1a, 1b, and 1c were prepared according to literature [33][34][35].NMR spectra were recorded on a Bruker Avance III 300 MHz FT NMR spectrometer (300.36MHz ( 1 H), 75.53 MHz ( 13 C)).Chemical shifts δ [ppm] are referenced to residual protonated solvent signals as internal standard CDCl 3 : δ = 7.26 ppm ( 1 H), 77.16 ppm ( 13 C).Elemental analyses were conducted on an Elementar Vario Micro Cube with CHN detector at 1200 °C.Results were found to be in good agreement (± 0.3%) with calculated values.Analytical thin layer chromatography (TLC) was performed on Merck silica gel 60-F 254 , and spots were visualized by UV light or by treatment with cerium ammonium molybdate solution or potassium permanganate.Column chromatography was performed using silica gel 60 Å from Acros Organics.
The rate constants were determined under pseudo firstorder conditions using UV-Vis spectroscopy.The decay in absorption of 1a at 545 nm or of 1c at 537 nm upon reaction with various amounts of the dienophiles 2a-2c was recorded.All reactions were conducted at 23 ± 0.5 °C using methanol in case of 1a and chloroform in case of 1c as the solvents.Such pseudo first-order rate constants were obtained by linearly fitting the logarithmized absorption over time data and the slope of the curve fit gave the respective observed rate constants for a given dienophile concentration, which varied between 0.05 and 0.1 mol dm −3 .The second-order rate constants were then derived from the linear regression equitation of the observed rate constants data plotted over the concentration of the dienophiles.

2-[3,6-Di(pyridin-2-yl)pyridazin-4-yl]
ethan-1-ol (3a) 500 mg 1a (2.12 mmol, 1.0 eq) and 0.192 cm 3 2a (2.54 mmol, 1.2 eq) were dissolved in 10 cm 3 toluene using a tightly closed Schlenk-tube and heated at 55 °C for 15 min whereupon the colour turned to pale yellow.The solvent was removed and the remaining solid was purified by column chromatography (CH 2 Cl 2 /MeOH 50/1) affording 413 mg (70%) 3a as a beige solid.NMR spectra were found to be in accordance with the ones described in Ref. [23].R f (CH 2 Cl 2 / MeOH 20/1) = 0.21; 1 H NMR (300 MHz, CDCl 3 , 25 °C): δ = 8.77-8.69(m, 2H, py 3,6 ), 8.67 (d, 1H, 3 J HH = 5.0 Hz, py 6 ), 8.60 (s, 1H, pz 5 ), 8.30 (d, 1H, 3 J HH = 8.0 Hz, py 3 ), 7.97 Table 1 Time needed for full conversion of 1a in the reaction with 2a in dependence of different reaction conditions Reaction conditions: to 50 mg 1a dissolved in 2 cm 3 of the respective solvent the corresponding amount of 2a was added and the reaction mixture was stirred at the given temperature in a tightly closed vessel; full conversion of 1a was defined as the time when the initially intense purple color of the reaction mixture vanished and the reaction mixture has turned pale yellow (visual inspection).The absence of 1a was confirmed by 1 H NMR spectroscopy, which in turn revealed the formation of side products, which were quantified in respect to the main product 3a