Kinetics and Mechanism of Oxidation of the Drug Intermediate 1-( 2-Hydroxyethyl ) piperidine by Bis ( hydrogenperiodato ) argentate ( III )

1-(2-Hidroxietil)piperidina (HEP) está presente em muitos fármacos e em protótipos de fármacos, mas seus mecanismos de oxidação são mal compreendidos. Assim, a reação de oxidação da HEP pelo bis(hidrogenoperiodato)-argentato(III) ([Ag(HIO6)2] ) foi investigada em meio alcalino aquoso. Piperidina e formaldeído foram identificados como principais produtos da reação através da técnica de espectrometria de massa com ionização electrospray. A cinética de oxidação foi acompanhada espectrofotometricamente no intervalo de temperatura de 25,0 a 40,0 oC, verificando-se que a reação é de primeira ordem em [Ag(III)] e de ordem fracionária em [HEP]. O estudo da dependência das constantes de velocidade kobsd em relação a [OHˉ] e [IO4ˉ] tot (concentração de periodato total) possibilitou estabelecer uma lei de velocidade e propor um mecanismo de reação. O mecanismo envolve a formação do complexo ternário periodato-Ag(III)HEP, seguido de sua decomposição a Ag(I) através de dois caminhos: um independente e outro facilitado por OH-. As constantes de velocidade e de equilíbrio bem como os parâmetros de ativação correspondentes às etapas determinantes de velocidade foram calculados.


Introduction
It is not exaggerated at all to describe 1-(2-hydroxyethyl)piperidine (HEP) as an important drug intermediate since it is involved in several clinical drugs and in many newly developed drug leads.For instance, raloxifene, 1,2 pipazethate 3,4 and flavoxate, 5,6 all containing HEP moiety, are currently prescribed drugs.Specifically, raloxifene has been used for treatment and prevention of osteoporosis in postmenopausal women, 1,2 whereas pipazethate is a non-narcotic antitussive drug. 3,4Flavoxate is an anticolinergic with antimuscarinic effects and has been employed to treat urinary bladder spasms. 5,6[9][10] On the other hand, HEP oxidation studies appeared to be very scarce, 11,12 providing very little mechanistic information.Thus, our group initiated a detailed spectrometric kinetic investigation on the oxidation of HEP by the bis(hydrogenperiodato)argentate(III) complex anion, [Ag(HIO 6 ) 2 ] 5-, establishing the rate law and delineating a mechanistic picture.
4][15][16][17] In the last few years, our investigations on [Ag(HIO 6 ) 2 ] 5− have led to two important findings: (i) the complex has shown some potential to modify peptides and drugs; 17,18 (ii) the Ag(III) complex can react with luminol or interact directly with some chemicals of analytical importance producing chemiluminescence.0][21][22][23][24] Meanwhile, we have also paid attention to the kinetic and mechanistic aspects of the Ag(III) oxidation reactions, gaining some insights on the possible oxidation mechanisms. 17,18,25,26In this work, we report our spectrometric, kinetic and mechanistic results on the oxidation of HEP by that Ag(III) complex.

Experimental
Instrumentation Electronic spectra and kinetic measurements were carried out on a UV-Vis spectrophotometer (TU-1901, Beijing Puxi, Inc., Beijing, China) equipped with several cell compartments whose temperature was controlled (± 0.2 o C) by circulating water from a thermostat (BG-chiller E10, Beijing Biotech, Inc., Beijing, China).Mass spectra were recorded on an Agilent 1200/6310 ion trap mass spectrometer equipped with an electrospray ionization (ESI) source.
The crystallized Na 5 [Ag III (HIO 6 ) 2 ] • 5H 2 O complex was synthesized using AgNO 3 , KIO 4 , K 2 S 2 O 8 , KOH and NaNO 3 according to the procedure reported by Balikungeri et al. 13 Electronic spectrum of aqueous Ag(III) complex solution displayed two absorption maxima at 253 and 362 nm, as previously reported in the literature. 14The stock solutions were prepared and measurement of the Ag(III) complex concentration was conducted by methods described elsewhere. 18All solutions were prepared with doubly distilled water.

Kinetic measurements
Kinetic measurements were carried out under pseudo first-order reaction conditions, [HEP] ≥ 10[Ag(III)], [IO 4 ˉ]tot ≥ 10[Ag(III)], where [IO 4 ˉ]tot denotes the total concentration of externally added periodate.Reaction was started by mixing equal volumes of the Ag(III) and HEP solutions, previously thermostated for at least 15 min, and the reaction mixture was quickly transferred to a pre-thermostated cell.Appropriate amounts of KOH, KIO 4 and KNO 3 were added to adjust [OHˉ], [IO 4 ˉ]tot and ionic strength (μ = 0.30 mol L -1 used throughout) of both Ag(III) and HEP solutions.
Figure 1 shows the time-resolved spectra for a set of reaction conditions, and the kinetic curves monitoring the absorbance at 360 nm as a function of time are given in Figure 1 as an insert (data point).The kinetic trace was fitted with single exponential equation (equation 1) using nonlinear least-squares routines, where A t , A 0 and A ∞ stand for the absorbances at time t, zero and infinity, respectively.
The excellent fitting, shown in Figure 1, demonstrates that the reaction is unequivocally first-order in [Ag(III)].The observed pseudo-first-order rate constants, k obsd , were calculated from the kinetic traces, and the average values of three independent runs are reported.Standard deviations of k obsd were usually less than 5%.

Test of free radical involvement
A volume of 50 mL of 0.10 mmol L -1 Ag(III) solution containing 0.10 mol L -1 KOH was mixed with 50 mL of 2 mmol L -1 HEP solution containing 8% acrylonitrile in a three neck flask, and stirred for 4 h in N 2 atmosphere.No precipitate of polyacrylonitrile was observed, suggesting that free radicals have not been formed in the reaction.Both solutions were purged for 20 min with nitrogen gas before the reaction.

Identification of the oxidation products
A volume of 100 mL of 1 mmol L -1 Ag(III) solution containing 0.1 mol L -1 KOH was mixed with an equal volume of 10 mmol L -1 HEP solution.After disappearance of the yellow color of the Ag(III) complex indicating the end of the reaction, the reaction mixture was divided in two equal parts.One was extracted with 200 mL of ether, washed once with water and then evaporated to dryness.The residue was dissolved in 2 mL of dilute HCl solution and subjected to mass spectrometry analysis.In the mass spectrum, one of the dominant peaks at m/z 86.4 was ascribed to protonated piperidine (piperidine•H + ), whose calculated m/z is 86.15.The minor difference is due to the calibration of the equipment.
An excess of KI and Na 2 S 2 O 3 was added to the second fraction of the reaction mixture, the pH was adjusted to 5-6 with 1 mol L -1 HCl.In this condition, the periodate was rapidly reduced, giving rise to I 3 ˉ which was quickly reduced to iodide by thiosulfite, 27 leading to precipitation of AgI and AgCl, after a while.These were filtered off and a saturated solution of DNPH (in 2 mol L -1 HCl) was added to the filtrate.The yellow precipitate, generated after aging the mixture for 2 days in a refrigerator, was washed with water, dissolved in DMSO and analyzed by mass spectrometry.The peak at m/z 233 (sodium peak), probably stems from the phenylhydrazone derivative formed by the reaction of DNPH with formaldehyde.Therefore, piperidine and formaldehyde probably are the oxidation products of HEP.

Influence of [HEP] on the oxidation rate
The oxidation rate as a function of [HEP] (0.60 ≤ [HEP] ≤ 6.0 mmol L -1 ) was studied at four temperatures and the plots of k obsd vs. [HEP] are displayed in Figure 2. Some simple calculations reveal that the reaction order n in [HEP] is fractional, namely 0 < n < 1.As a matter of fact, good linear relationships between 1/k obsd and 1/[HEP] were found at all the temperatures (not shown), suggesting that they may follow a Michaelis-Menten type kinetics of enzyme-catalyzed reactions. 28It implies that HEP forms an intermediate complex with Ag(III).
The time-resolved spectra in Figure 1 show an absorption peak at 360 nm, which is in contrast with 362 nm peak of the Ag(III) complex in pure water.This blue shift of the band suggests that there is an interaction between Ag(III) and HEP, supporting the formation of a complex between Ag(III) and HEP (vide infra).

Influence of [OHˉ] on the oxidation rate
The influence of [OHˉ] on the reaction rate was studied in the 0.020 ≤ [OHˉ] ≤ 0.20 mol L -1 range, and the plots of k obsd vs.
[OHˉ] at different temperatures are shown in Figure 3.The linear plots afforded well-defined and significantly positive intercepts.Similar relationships between k obsd and [OHˉ] were observed for the oxidation reactions of (L)-serine and (L)-threonine 18 and mephenesin 17 with Ag(III).But, analogous reactions with (L)-proline and (DL)-pipecolinate 25,26 showed that k obsd is essentially independent of [OHˉ].

Oxidation mechanism
Provided [Ag(HIO 6 )(OH)(H 2 O)] 2-is the reactive species, it should be the intermediate complex formed in the reaction of Ag(III) and HEP, such that the simple expression k obsd = a + b[OHˉ] can describe the k obsd dependency on [OHˉ] shown in Figure 3. Also, this two-term expression reveals that the oxidation reaction may take place through two channels: one independent of OHˉ and another facilitated by that same species.As a consequence, the reaction mechanism shown in Scheme 1 was proposed considering the above observations and hypothesis.
In this mechanism, the intermediate is proposed to be a ternary complex (structure shown in Scheme 1) formed between [Ag(HIO 6 )(OH)(H 2 O)] 2-and HEP in a pre-equilibrium step.This decomposes to Ag(I) according to two parallel rate-determining steps (described by k 1 and k 2 , respectively).
The rate expression deduced in terms of total concentration of silver(III), [Ag(III)] tot , where [H 2 IO 6 3-] e denotes the equilibrium concentration of H 2 IO 6 3-, is shown in equation 3.
(3) Hence, (4)   The kinetic measurements were carried out fulfilling the condition of [HEP] ≥ 20 [Ag(III)] tot .As a consequence, only a small fraction of HEP is consumed in the formation of ternary complex 3 in Scheme 1.That is estimated to be only 1-3% of the total HEP (less than kinetic standard deviations) such that [HEP] is essentially the total concentration of HEP in equations 3 and 4.
Periodate speciation in alkaline medium was discussed earlier, 17,18 suggesting that H 2 IO 6 3-is the predominant species in the studied [OHˉ] range. 17,18[H 2 IO 6 3-] e as a fraction of [IO 4 ˉ]tot can be expressed by equations 5 and 6, where β 2 and β 3 are known equilibrium constants of periodate in alkaline medium. 29,30The periodate speciation, the equilibrium constants β 2 and β 3 , and more detailed discussions on those parameters are referred to our earlier In addition, equation 7 can be derived substituting equation 5 into equation 4: (7)   Rate and equilibrium constants The pre-equilibrium constant K 1 (equation 7) as a function of temperature was previously derived 18 and is virtually constant in the studied temperature range.Equation 7 was employed to fit the k obsd vs. [HEP] data in Figure 2 using a nonlinear least-square method with K 2 and (k 1 + k 2 [OHˉ]) as adjustable parameters, and K 1 = 6.0 × 10 −4 mol L -1 and f([OHˉ]) = 0.958 as constants.The fittings were very good (see Figure 2), and K 2 values derived from the fittings are listed in Table 2.
Since all rate and equilibrium constants in equation 7 were derived, it is possible to predict or re-calculate the k obsd using the same equation.In this way, k obsd was predicted as a function of [IO 4 − ] tot and the values given in Table 1.Obviously, the predicted values are in good agreement with the measured ones when the experimental errors are taken Here, it is necessary to stress that the measured k obsd values (Table 1) were not used in the derivation of the rate and equilibrium constants.As a consequence, the good agreement between the measured and predicted k obsd values strongly support the rate law, indicating that the proposed reaction mechanism is reasonable.

Activation parameters
The study of the temperature dependency of k 1 and k 2 enabled the calculation of the corresponding activation parameters.The activation enthalpies and entropies were determined from the Eyring plots shown in Figure 5 (Table 2).The ∆H 2 ‡ = 43 ± 3 kJ mol −1 and ∆S 2 ‡ = -134 ± 9 J K −1 mol −1 are typical for a bimolecular reaction.However, k 1 is still regarded as the first-order rate constant by convention because water concentration is constant, where the bimolecular nature is accounted for by the negative value of ∆S 1 ‡ .In fact, a one step two-electron transfer process in the rate-determining steps is in line with our free radical involvement test.The reaction intermediate, namely aminium cation in Scheme 1, generated from HEP in the rate-determining steps, is similar to those proposed when ternary amines are oxidized by hypochlorite. 31

Conclusions
The reaction products and kinetics for the oxidation of the important drug component HEP by an Ag(III) complex were analyzed by spectrometric techniques.A reaction mechanism involving the indirect formation of a periodato-Ag(III)-HEP ternary complex as an intermediate is supported by both the time-resolved spectra and rate dependence on [HEP].The ternary complex decomposes via two reaction channels as the rate-determining steps.The equilibrium constant and rate constants, and their corresponding activation parameters, were evaluated, and a consistent mechanism proposed, enabling the in-depth understanding of the oxidation process.This might be a reference mechanism for the oxidative degradation of the HEP-containing drugs.

Table 2 .
Derived values of equilibrium constant K 2 , rate constants k 1 and k 2 , and activation parameters