An unusual double radical homolysis mechanism for the unexpected activation of the aldoxime nerve-agent antidotes by polyhalogenated quinoid carcinogens under normal physiological conditions

Author’s Accepted Manuscript An Unusual Double Radical Homolysis Mechanism for the Unexpected Activation of the Aldoxime Nerve-Agent Antidotes by Polyhalogenated Quinoid Carcinogens under Normal Physiological Conditions Lin-Na Xie, Jie Shao, Chun-Hua Huang, Feng Li, Dan Xu, Balaraman Kalyanaraman, Ben-Zhan Zhu

PII: DOI: Reference:

www.elsevier.com

S0891-5849(18)32220-2 https://doi.org/10.1016/j.freeradbiomed.2018.10.425 FRB13989

To appear in: Free Radical Biology and Medicine Received date: 1 May 2018 Revised date: 12 October 2018 Accepted date: 13 October 2018 Cite this article as: Lin-Na Xie, Jie Shao, Chun-Hua Huang, Feng Li, Dan Xu, Balaraman Kalyanaraman and Ben-Zhan Zhu, An Unusual Double Radical Homolysis Mechanism for the Unexpected Activation of the Aldoxime NerveAgent Antidotes by Polyhalogenated Quinoid Carcinogens under Normal Physiological Conditions, Free Radical Biology and Medicine, https://doi.org/10.1016/j.freeradbiomed.2018.10.425 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

An Unusual Double Radical Homolysis Mechanism for the Unexpected Activation of the Aldoxime Nerve-Agent Antidotes by Polyhalogenated Quinoid Carcinogens under Normal Physiological Conditions Lin-Na Xie1,2, Jie Shao1,2, Chun-Hua Huang1,2, Feng Li1,2, Dan Xu1,2, Balaraman Kalyanaraman3, Ben-Zhan Zhu1,2,* 1

State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research

Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, P. R. China 2

University of Chinese Academy of Sciences, Beijing 100049, P. R. China; 3Medical

College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226-0509, USA *Corresponding author. State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, The Chinese Academy of Sciences. P.O. Box 2871. Beijing, P. R. China 100085. Phone: 86-10-62849030; Fax: 86-10-62923563. [email protected] Abstract: We have recently shown that the pyridinium aldoximes, best-known as therapeutic antidotes for chemical warfare nerve-agents, could markedly detoxify the carcinogenic tetrachloro-1,4-benzoquinone (TCBQ) via an unusual double Beckmann fragmentation mechanism. However, it is still not clear why pralidoxime (2-PAM) cannot provide full protection against TCBQ-induced biological damages even when 2-PAM was in excess. Here we show, unexpectedly, that TCBQ can also activate pralidoxime to generate a reactive iminyl radical intermediate in two-consecutive steps, which was detected and unequivocally characterized by the complementary application of ESR spin-trapping, HPLC/MS and nitrogen-15 isotope-labeling studies. The same iminyl radical was observed when TCBQ was substituted by other halogenated quinones. The end product of iminyl radical was isolated and identified as its corresponding reactive and toxic aldehyde. Based on these data, we proposed that the reaction of 2-PAM and TCBQ might be through the following two competing

pathways: a nucleophilic attack of 2-PAM on TCBQ forms an unstable transient intermediate, which can decompose not only heterolytically to form 2-CMP via double Beckmann fragmentation, but also homolytically leading to the formation of a reactive iminyl radical in double-steps, which then via H abstraction and further hydrolyzation to form its corresponding more toxic aldehyde. Analogous radical homolysis mechanism was observed with other halogenated quinones and pyridinium aldoximes. This study represents the first detection and identification of reactive iminyl radical intermediates produced under normal physiological conditions, which provides direct experimental evidence to explain only the partial protection by 2-PAM against TCBQ-induced biological damages, and also the potential side-toxic effects induced by 2-PAM and other pyridinium aldoxime nerve-agent antidotes. Graphic Abstract fx1 Key Words: Radical homolysis mechanism; Iminyl radical; Tetrachloro-1, 4-benzoquinone

(TCBQ);

Pyridinium

aldoximes;

Nerve-agent

antidotes;

Polyhalogenated quinoid carcinogens

Introduction Polyhalogenated quinones are a group of reactive intermediates which can cause various harmful consequences in vivo, such as acute nephrotoxicity, hepatoxicity, and carcinogenesis [1, 2].. Pentachlorophenol (PCP) has been extensively used for wood preservation. In addition, PCP was used to prevent snail fever as a snail-killing compound in China and other developing countries. The U.S. EPA has identified that at least 20% of National Priorities List sites are contaminated with PCP, and the International Association for Research on Cancer (IARC) has recently considered PCP as a Group-I environmental carcinogen [3, 4].. During the oxidation and destruction processes for PCP and other polychlorinated aromatic compounds via different chemical and enzymatic systems, tetrachloro-1,4-benzoquinone (TCBQ) has been generated as a reactive and transient intermediate or final product [5-8].. Recently, TCBQ was detected as one of many disinfection quinoid byproducts in

chlorination of drinking and swimming pool water [9, 10].. TCBQ was often used as a powerful oxidizing or dehydrating agent in organic synthesis. Pyridinium aldoximes have been used as acetylcholinesterase activators since the 1950s for the medical treatment of poisoning with organophosphorus nerve agents and pesticides. More than 300,000 fatalities every year are due to organophosphorus pesticide poisoning worldwide, resulting in a serious public health problem[11].. In addition, organophosphorus nerve agents remain a persistent threat to the public due to terrorist attacks and armed conflicts. Aldoximes possess high nucleophilicity that can displace the phosphyl group from the catalytic serine to restore catalytic activity of the acetylcholine esterase (AchE) [12-16].. In recent years, researchers have devoted enormous efforts to the synthesis and development of new compounds as potential antidotes. Only four pyridinium aldoximes (i.e., pralidoxime or 2-PAM, trimedoxime, obidoxime, and HI-6), are in clinical use [13, 17, 18].. Pralidoxime and obidoxime are the two currently clinically used drugs, and trimedoxime (TMB-4) and HI-6 have been tested in humans and are available for human use in some countries[12].. During our investigations on the reaction mechanisms between hydroxlamines and polyhalogenated quinoid compounds[19-21].. We found recently that pyridinium aldoximes can detoxify TCBQ and other polyhalogenated quinoid compounds[22].. This detoxification was found to be via an unusually mild and facile Beckmann-type double fragmentation mechanism. Further, we found that 2-PAM could provide remarkable, but not full protection against TCBQ/H2O2-induced DNA double-strand breaks in isolated DNA even when 2-PAM was in excess. So we hypothesized that there might be other pathway for the reaction between TCBQ and 2-PAM except for Beckmann fragmentation. The molecular mechanism for the reaction between TCBQ and pyridinium aldoximes was then extensively examined using ESR spin-trapping, HPLC/MS and nitrogen-15 isotope-labeling methods. Unexpectedly, we found that TCBQ could also activate 2-PAM to form a reactive nitrogen-centered iminyl radical intermediate via an unusual double radical homolysis pathway.

Results and Discussion An unusual new nitrogen-centered radical was produced by the reaction between TCBQ and 2-PAM in two consecutive steps In our previous studies, we found that TCBQ and other polyhaloquinoid compounds can produce highly reactive hydroxyl radical (•OH), carbon-centered quinone ketoxy radical and alkoxyl radical in the presence of H2O2 and organic hydroperoxides through a metal-independent mechanism [3, 23-29]., which can lead to DNA strand breakage, base oxidations, cyto- and genotoxicity [30-34].. Recently we found that 2-PAM can provide strong protection against TCBQ/H2O2-induced DNA double-strand breaks in isolated DNA (Fig. S1). The protection by 2-PAM was proposed to be mainly due to its efficient inhibition of •OH formation by TCBQ/H2O2 via double Beckmann fragmentation. However, we found that 2-PAM can’t provide complete protection against DNA damage, even when the added 2-PAM was in excess. Therefore, we hypothesized that there might be other pathway between TCBQ and 2-PAM except for Beckmann fragmentation. In order to test whether it was a free radical pathway, the ESR spin-trapping method was adopted in TCBQ/H2O2/2-PAM reaction system using 5, 5-dimethyl-1-pyrroline N-oxide (DMPO), which is a classic trapping agent for oxygen-, carbon-, and nitrogen-centred radical [35-38].. The intensity of typical 4-line DMPO/•OH signal was found to decrease progressively as the addition of 2-PAM increased (Fig. 1). To our surprise, the DMPO/•OH radical signal didn’t completely disappear in the presence of excess 2-PAM, and a new ESR signal was observed in addition to the typical 4-line DMPO/•OH signal (Fig. 1). We further found that the new ESR signal was a special 18-line signal, which can be produced by TCBQ/2-PAM even in the absence of H2O2. Further simulation studies suggested that the new radical might be a nitrogen-centered radical (Fig. 1). The nitrogen-centered radical was produced only by the combination of TCBQ and 2-PAM together, but not by either of them alone. The formation of the nitrogen-centered radical was found to be time and dose-dependent on the molar ratios of 2-PAM/TCBQ. The time courses of formation and decay of the ESR signal

of the nitrogen-centered radical increased as the molar ratio increased (Fig. 2 and Fig. S2F). The influence of pH on the formation and decay of the ESR signal of nitrogen-centered radical was also studied. We found that nitrogen-centered radical cannot be observed at pH = 2; as the pH increased, time courses of formation and decay of nitrogen-centered radical were both accelerated progressively; at pH = 5.0, the duration time of DMPO-nitrogen-centered radical adduct was relatively long, and its intensity was relatively high, which indicated that DMPO-nitrogen-centered radical adduct was most stable at pH = 5.0; as the pH ≥ 6.0, both formation and decay rate of DMPO-nitrogen-centered radical adduct continued increasing, and the decay rate was faster than the formation rate, so we could only observe the decay process at the pH ≥ 9.0 (Fig. S3). In order to mimic a standard physiological condition, we chose pH 7.4 as the condition in this study. 2-PAM : TCBQ : H2O2 0:1:10

0:1:0

1:1:10 4:1:10 ???

?? ??

?? ??

???

10:1:10 100:1:10

100:0:10 100:1:0 simulation

Fig. 1: ESR spectra of DMPO/N-centered radical produced by TCBQ/2-PAM in the presence or absence of H2O2. Reactions were all carried out at 20 ℃ for 2 min in Chelex-treated phosphate buffer (100 mM, pH 7.4), DMPO: 100 mM; TCBQ: 0.1mM; H2O2, 1 mM. The central signal in the spectra was identified as tetrachloro-1,4-semiquinone anion radical (TCSQ•-) (g=2.0058). The hyperfine

splitting constants for the DMPO-nitrogen radical: aH=16.30 G, aN1=2.75 G, aN2=14.90 G. The hyperfine splitting constants for the DMPO-hydroxyl radical: aH = aN = 14.89 G. The experiment was repeated twice.

2-PAM:TCBQ 50:1 10:1

4:1 1:1

1:10 1:50

Fig. 2: Formation of DMPO/N-centered radical by interactions between TCBQ and 2-PAM. The reaction was conducted in Chelex-treated phosphate buffer (100 mM, pH 7.4) after 1 minute. DMPO: 100 mM, TCBQ:2-PAM = 1:1 = 0.05 mM:0.05 mM. The experiment was repeated twice.

In our previous research, we found that TCBQ and 2-PAM could form the initial hydroxylation intermediate trichlorohydroxy-1,4-benzoquinone (TrCBQ-OH), and then very slowly further to the final 2,5-dichloro-3,6-dihydroxy-1,4-benzoquinone (DDBQ) at neutral pH through two consecutive Beckmann fragmentation. To test whether TrCBQ-OH can also react with 2-PAM to produce the same nitrogen-centered radical, TrCBQ-OH was synthesized as previously described [25].. We found that this was indeed the case (Fig. 3A), suggesting that the radical produced by the second step is independent of the first step. The time courses of formation and decay of the ESR signal of the nitrogen-centered radical was found to depend on the molar ratio of TrCBQ-OH/2-PAM (Fig. 3B). Even though higher 2-PAM ratios results in higher signal intensity of nitrogen-centered radical, excessive 2-PAM can

also reduce nitrogen-centered radical via hydrogen abstraction from 2-PAM and increase the rate of signal loss (Fig. 3B). These results indicate clearly that the interaction between TCBQ and 2-PAM could produce a special nitrogen-centered radical intermediate in two consecutive steps.

TrCBQ-OH : 2-PAM = 1 : 1

Signal Intensity (x 103)

A

TrCBQ-OH : 2-PAM = 1 : 2

TrCBQ-OH : 2-PAM = 1 : 4

TrCBQ-OH : 2-PAM = 1 : 10

B TrCBQ-OH + 2-PAM

8

TrCBQ-OH : 2-PAM 6

1/1 1/2 1/4 1/10

4

2 0

5

10

15

20

25

30

35

Time (min)

Fig. 3: The time course of ESR signal intensity of N-centered radicals produced by 2-PAM and TrCBQ-OH. (A) ESR spectra of DMPO/nitrogen-centered radical produced by TrCBQ-OH and 2-PAM. (B) Time course of ESR signal intensity of nitrogen-centered radical produced by TrCBQ-OH and 2-PAM. The reaction was conducted in Chelex-treated phosphate buffer (100 mM, pH 7.4) at room temperature, and the reaction time was 1 min. TrCBQ-OH: 0.05 mM; DMPO: 100 mM. Each point was repeated twice, and the SD was less than 5%.

The new radical intermediate was identified as iminyl radical by ESR spin-trapping, nitrogen-15 isotope-labeling and HPLC/MS studies According to our previously proposed mechanism for dramatic acceleration of TCBQ hydrolysis by 2-PAM (Scheme S1), a nucleophilic attack of 2-PAM on TCBQ initially forms an unstable transient intermediate, which undergoes a Beckmann-type fragmentation. Since the N-O bond of the intermediate, an oxime derivative, is usually comparatively weak, we hypothesized that the unstable transient intermediate might also undergo a homolytical decomposition leading to the formation of an N-centered radical. If the above hypothesis were right,

15

N-labeled 2-PAM

([15N]-2-PAM) should generate a different signal with TCBQ as the nuclear-spin quantum number of

15

N is different from that of

14

N (the nuclear-spin quantum

number of 15N and 14N is 1/2 and 1, respectively). To test whether this is the case, the same experiments were conducted with synthesized [15N]-2-PAM. Indeed, we found that the ESR spectrum of the N-centered radical turned into a 12-line signal (Fig. 4), which unequivocally confirms that the radical was an N-centered iminyl radical.

A TCBQ/2-PAM B TCBQ/[15N]-2-PAM

Fig. 4: The new radical intermediate was further identified as iminyl radical by ESR spin-trapping with nitrogen-15 isotope-labeling study. The reaction was conducted in Chelex-treated phosphate buffer (100 mM, pH 7.4) after 1 minute. DMPO: 100 mM, TCBQ, 0.05 mM; 2-PAM, 5.0 mM, [15N]-2-PAM: 5.0 mM. The hyperfine splitting constants for the DMPO-[15N]-iminyl radical: aH = 16.62 G, aN1 = 3.83 G, aN2 = 14.91 G. The experiment was repeated twice.

The formation of DMPO-iminyl radical adduct was further investigated by HPLC-ESI-Q-TOF-MS. In the presence of DMPO, a new peak with the retention time at 3.93 min was observed at m/z 233 from the reaction between TCBQ and 2-PAM. The MS/MS spectrum (ESI-positive) showed that the new compound was a DMPO adduct with the iminyl radical (MW. 120) (for simplicity, this DMPO adduct was referred as DMPO-120) (Fig. 5A-5C, Fig S4A, Fig S4B). Besides, when DMPO was substituted

by

5-t-butoxycarbonyl-5-methyl-1-pyrroline

N-oxide

(BMPO),

a

spin-trapping agent with similar chemical structure like DMPO, a BMPO-120 adduct was also observed (Fig. 5D-5F, Fig S4C, Fig S4D). These results further confirmed the formation of iminyl radical produced by TCBQ and 2-PAM.

Relative Intensity

Cl N

N OH

N

DMPO-120 0

2

100

4

6 Time (min)

8

10

But O

N N OH

O

N

BMPO-120

100

5

10 Time (min)

15

20

319

E

%

% 150

180

C

210

240

m/z 300

270

114

100

120 93

80

100

275

300

F

325

350

m/z 400

375

144

120

119 120

250

%

%

100

12.96 Cl

0

12

233

B

D

Relative Intensity

3.84

A

200

233 m/z 140

160

180

200

220

240

50

100

150

200

319 250

300

m/z 350

G HPLC/MS/MS Fragmentation Analysis 114

N N OH 120

200

Cl N

But O N O

Cl N

OH 120 HO

N

144

Fig. 5: Isolation and identification of the relatively stable DMPO-iminyl radical and BMPO-iminyl radical produced by TCBQ/2-PAM in the presence of DMPO or BMPO, respectively. (A, D)The HPLC/MS selected ion monitoring (SIM）profile of the reaction mixtures of 2-PAM with TCBQ in CH3COONH4 buffer (pH 7.4, 0.1 M). MS spectrum of DMPO-iminyl radical at the retention time of 3.84 min in the HPLC/MS chromatogram (B) and its MS/MS spectrum (C). MS spectrum of BMPO-iminyl radical at the retention time of 12.96 min in the HPLC/MS chromatogram (E) and its MS/MS spectrum (F). (G) HPLC/MS/MS fragmentation analysis of DMPO-120 and BMPO-120. The reaction was conducted at the ratio of 1:100 (TCBQ/2-PAM) in CH3COONH4 buffer (pH 7.4, 0.1 M) at room temperature, and the reaction time was 2 min. TCBQ, 0.05 mM; 2-PAM, 5.0 mM. The experiment was repeated twice.

Analogous iminyl radicals could be produced by other halogenated quinones and aldoximes antidotes We found that not only TCBQ, but also other chlorinated benzoquinones (CnBQs) could produce the same iminyl radical with 2-PAM (Fig. 6). These include 2-chloro-1,4-benzoquinone

(2-CBQ),

2,3-dichloro-

(2,3-DCBQ),

2,5-dichloro-

(2,5-DCBQ), 2,6-dichloro- (2,6-DCBQ), and 2,3,5-trichloro- (TrCBQ). The formation of iminyl radical was found to be dependent on the molar ratios of 2-PAM/CnBQs: the more for added 2-PAM, the stronger for ESR signal of iminyl radical, and the faster for the reaction. From these results, we found that the order for the iminyl radical signal intensity produced by the six CnBQs is: TCBQ ≈ 2,3-DCBQ ＞ TrCBQ ＞ 2,5-DCBQ ＞ 2,6-DCBQ ＞ 2-CBQ (Fig. S2). Moreover, the same DMPO-120 adduct was also observed by HPLC-ESI-Q-TOF-MS when TCBQ was substituted by other CnBQs (Fig. S5).

TCBQ / 2-PAM TrCBQ / 2-PAM 2,3-DCBQ / 2-PAM 2,5-DCBQ / 2-PAM

2,6-DCBQ / 2-PAM 2-CBQ / 2-PAM Fig. 6: Analogous iminyl radical could be produced by other halogenated quinones and 2-PAM. The reaction was conducted in Chelex-treated phosphate buffer (100 mM, pH 7.4) after 1 minute. DMPO: 100 mM, CnBQ: 0.05 mM, 2-PAM: 5 mM. The hyperfine splitting constants for the DMPO-iminyl radical by different CnBQs/2-PAM: aH=16.30 G, aN1=2.75 G, aN2=14.90 G. The experiment was repeated twice.

Further, we found that the same iminyl radical can also be produced by 2-PAM and other halogenated quinones such as 2,5-dibromo- (2,5-DBrBQ), tetrafluoro-

(TFBQ) and tetrabromo-1,4-benzoquinone (TBrBQ), tetrachloro-1,2-benzoquinone (O-TCBQ), and tetrachlorohydroquinone (TCHQ), but not by tetrachloro-catechol (TCC),

1,4-benzoquinone

(BQ),

2-methyl-1,4-benzoquinone

(2-MeBQ)

and

tetramethyl-1,4-benzoquinone (TMeBQ) (Fig. S6). These results clearly indicated the importance of halogen substitution of halogenated quinones played in the radical hemolysis reaction, because halogen can increase the electrophilicity of the bonded carbon, then the intermediates can be readily formed by the nucleophilic attack of the pralidoxime anion (Py+(CH3)−CH=N−O−) on the halogen-substituted position on polyhalogenated quinones. If there is no halogen substitution in quinones, like BQ, 2-MeBQ, and TMeBQ, it would be much difficult to form substitution intermediates, so the radical homolysis would not take place in non-halogenated quinones. Moreover, similar iminyl radicals can also be produced when 2-PAM was substituted by other aldoximes antidotes such as HI-6, obidoxime, and TMB-4 (Fig. 7). These results indicated that the production of iminyl radical is a general phenomenon for all halogenated quinones and all aldoximes antidotes.

TCBQ : 2-PAM = 1 : 50, 1.2 min TCBQ : HI-6 = 1 : 100, 1.3 min TCBQ : Obidoxime = 1 : 10, 0.9 min

TCBQ : TMB-4 = 1 : 20, 1.2 min

Fig. 7: Analogous iminyl radicals could be produced by TCBQ and other aldoximes antidotes. The ESR specturm of DMPO-nitrogen radical, the reaction was conducted in Chelex-treated phosphate buffer (100 mM, pH 7.4) at room temperature. DMPO: 100 mM, TCBQ: 0.05 mM. The hyperfine splitting constants for the DMPO-iminyl

radical produced by TCBQ/HI-6: aH = 16.08 G, aN1 = 2.65 G, aN2 = 14.86 G. The hyperfine

splitting

constants

for

the

DMPO-iminyl

radical

produced

by

TCBQ/Obidoxime: aH = 17.12 G, aN1 = 2.59 G, aN2 = 14.91 G. The hyperfine splitting constants for the DMPO-iminyl radical produced by TCBQ/TMB-4: aH = 17.39 G, aN1 = 2.59 G, aN2 = 14.92 G. The experiment was repeated twice.

The end product of iminyl radical was identified as its corresponding aldehyde, N-methyl-2-pyridinecarboxaldehyde chloride As reported, there are mainly three pathways for the conversion of iminyl radicals [39].. The first one is that iminyl radicals end up as imines after H atom abstraction from solvent, which can further hydrolyze to their corresponding aldehydes or ketones. Secondly, the iminyl radicals may also be converted into N-heterocycles, such as indazoles or quinazolines through ring closure, which is usually used for chemical synthesis. Thirdly, the iminyl radicals terminate rapidly by N to N coupling to give azines. In

our

study

of

2-PAM

and

CnBQs,

only

the

aldehyde

N-methyl-2-pyridinecarboxaldehyde chloride (N-Me-2-PCA) was observed; in other words, the iminyl radical transformed to aldehyde via the first pathway. We met with much difficulty in isolating N-Me-2-PCA (the reactant 2-PAM) and the Beckmann fragmentation product 2-cyano-1-methylpyridinium chloride (2-CMP). After quite a long time, we eventually found a method to isolate and quantify N-Me-2-PCA using HPLC-ESI-TSQ-MS. The reaction rate between TCBQ and 2-PAM is relatively fast. When 2-PAM was consumed completely after 36 min, the yield of N-Me-2-PCA reached 9.0%, and the Beckmann fragmentation product 2-CMP was 90.0% (Fig. 8A).

0.015

A TCBQ + 2-PAM

Concentration (mM)

0.05

B N-Me-2-PCA 0.012

0.04 2-CMP N-Me-2-PCA 2-PAM

0.03

0.009

0.02

0.006

0.01

0.003

0.00

0.000 0

5

10

15

20

25

30

35

40

Time (min)

2,6-DCBQ 2,5-DCBQ 2-CBQ 2,3-DCBQ TrCBQ TCBQ

0

20

40

60

80

100

Time (min)

Fig. 8: The quantification of 2-CMP, N-Me-2-PCA and 2-PAM produced by 2-PAM and TCBQ, and the comparison of the yield of N-Me-2-PCA produced by 2-PAM and different chlorinated benzoquinones. (A) The concentration of 2-PAM markedly decreased in the addition of excess TCBQ, along with the generation of 2-CMP and N-Me-2-PCA. (B) The quantification of N-Me-2-PCA in different CnBQs/2-PAM. The reaction was conducted in Chelex-treated CH3COONH4 buffer (10 mM, pH 7.4) at room temperature. CnBQs: 0.20 mM, 2-PAM: 0.05 mM. Each point was repeated twice, and the SD was less than 5%.

As 2-PAM could only partially protect TCBQ/H2O2-induced DNA damage, and the radical homolysis product of 2-PAM/TCBQ is N-Me-2-PCA. Then the question is whether N-Me-2-PCA is more toxic than 2-PAM? To answer this question, the cytotoxicity of N-Me-2-PCA was measured by MTT assay in HepG2 cells (Fig. S8), and the IC50 was calculated to be 56.79 mM, which indicated that N-Me-2-PCA is indeed more toxic than 2-PAM (141.01 mM). As the literatures reported[40, 41]., reactive nitrogen species (RNS) can cause biological damage, so we speculated that the iminyl radical is also reactive and toxic, it could not only go through hydrogen abstraction, but also form DNA-adduct or lead to DNA damage, which can well explain why 2-PAM provided only partial protection against TCBQ/H2O2-induced biological damages in isolated DNA and cells. Moreover, we also detected and quantified the yields of N-Me-2-PCA and 2-CMP produced by other CnBQs and 2-PAM (Fig. 6B and Fig. S9). In

2,6-DCBQ/2-PAM and 2-CBQ/2-PAM, the yield of 2-CMP was just a little more than that of N-Me-2-PCA, and the reaction rate was comparatively slower. However, for the other four CnBQs (TCBQ, TrCBQ, 2,3-DCBQ, 2,5-DCBQ) and 2-PAM, the yield of 2-CMP was much higher than that of N-Me-2-PCA, and the reaction rate was much faster.

An unusual double radical homolysis mechanism for the activation of 2-PAM by TCBQ During our investigation of the reaction mechanism between CnBQs and 2-PAM, we accidentally observed the formation of the special iminyl radical intermediate, suggesting that the molecular mechanism for the reaction is not only Beckmann fragmentation, but also radical homolysis: A nucleophilic reaction may take place between TCBQ and pralidoxime anion (Py+(CH3)-CH=N-O-), forming an unstable intermediate Py+(CH3)-CH=N-O-trichloro-1,4-benzoquinone, which either undergoes Beckmann fragmentation to convert the highly toxic TCBQ and 2-PAM to generate the much less toxic DDBQ and the non-toxic 2-CMP, respectively; or decompose homolytically to produce an iminyl radical and TrCBQ-O•. The iminyl radical ends up as aldehyde (N-Me-2-PCA) after H atom abstraction and further hydrolyzation (Scheme 1). TrCBQ-O• can disproportionate to form the ionic form of TrCBQ-OH. In the presence of excess 2-PAM, TrCBQ-OH may further react with 2-PAM to form analogous reaction intermediates, which can further decompose to the iminyl radical and convert to another molecule of N-Me-2-PCA via a second-step radical homolysis reaction. To our knowledge, this is the first report that CnBQs and 2-PAM can produce the iminyl radical via a radical homolysis mechanism under normal physiological conditions.

O Cl Cl

O

Cl

N

N H

O

TCBQ

O

Nucleophilic Substitution

Cl

Cl

Cl

Cl

O

Cl

N

O

2-CMP

N

TrCBQ-O

H

I

2-PAM

1st Beckmann Fragmentation

1st Radical Homolysis

O

O Cl

Cl

Cl

O

Dismutation

Cl

Cl

Cl

O

N O

TrCBQ-O

TrCBQ-O

H Abstraction H2O

HN

2-PAM

Cl

N H

O

N H

O

N H

N-Me-2-PCA H Abstraction

2-CMP

2nd Beckmann Fragmentation

H N

O O

Cl

Cl

O

N

DDBQ II

2nd Radical Homolysis

O HO

Cl

Cl

O

N

O

O

N H

DDBQ-O

Scheme 1: Proposed two competing pathways for the reaction between TCBQ and 2-PAM under normal physiological conditions: Double radical homolysis vs double Beckmann fragmentation

Possible biological and environmental relevance We found that this unusual radical homolysis reaction mechanism is not only limited to TCBQ and 2-PAM, but it is also a general mechanism for all halogenated quinoid compounds and aldoximes antidotes. Therefore, our findings may have interesting chemical, biological and environmental implications. Iminyl radicals are important and promising intermediates to be used in organic synthesis. In the previously reported studies, the methods used to produce iminyl radicals include visible-light promoting, UV photolysis and microwave-assistant decomposition of oxime esters or oxime ethers [42-47].. In this study, we found that TCBQ and other halogenated quinones could act as new activating agents for aldoximes to produce iminyl radical. To our knowledge, this is the first detection and identification of iminyl radical under neutral pH and at room temperature, and this research provides a new way to produce iminyl radicals by halogenated quinones and aldoximes under normal physiological conditions.

Based on above results, there are two competing pathways for TCBQ/2-PAM reaction: one is detoxification via Beckman fragmentation; the other is activation via radical homolysis to form reactive iminyl radical and toxic aldehyde. Comparing the yields of end products of two pathways, we found that in both 2,6-DCBQ/2-PAM and 2-CBQ/2-PAM systems, the reactions undergo through the two pathways evenly; with other CnBQs (TCBQ, TrCBQ, 2,3-DCBQ, 2,5-DCBQ) and 2-PAM, the Beckman fragmentation is the major pathway. This study represents a universal radical homolysis mechanism for all halogenated quinonoid compounds and aldoximes antidotes, which complements our previously proposed Beckmann fragmentation mechanism. Since TCBQ and 2-PAM can produce the reactive iminyl radical and the toxic aldehyde, it can well explain why 2-PAM provided only partial protection against TCBQ/H2O2-induced biological damages in isolated DNA and cells. In fact, four pyridinium aldoximes (pralidoxime, trimedoxime, obidoxime, and HI-6) have been employed for clinical treatment of the poisonings by organophosphorus pesticides or nerve-agents [15].. As demonstrated in our recent study, these compounds are also effective for detoxification of polyhaloquinones, which are a class of carcinogenic intermediates and recently identified disinfection byproducts, via Beckmann-type fragmentation reactions. Interestingly, pyridinium aldoximes react directly with organophosphorus chemical warfare agents, or activate the inhibited enzyme by forming phosphyloximes, which decompose to produce phosphilic acid and the corresponding nitriles in vitro, possibly via an analogous Beckmann-type fragmentation reaction [12, 16].. But it is worth noting that there are adverse reactions associated with the clinical use of pyridinium aldoximes for the treatment of organophosphorus poisonings. The rapid injection of pyridinium aldoximes can produce headache, disturbance of vision, nausea, dizziness, tachycardia and increased blood, hyperventilation and even hepatic dysfunction after multiple dosing [48-50].. From this study, we speculate that some of the potential side toxic effects of pyridinium aldoximes after treating nerve agent intoxication, might be possibly due to the formation of the reactive iminyl radicals and/or toxic aldehyde by decomposition of the conjugates formed between nerve agents and pyridinium

aldoximes, possibly via an analogous homolysis mechanism. Further investigations are needed to study this hypothesis.

Materials and Methods Materials Tetrachloro-1,4-benzoquinone (TCBQ; tetrachloro-p-benzoquinone or p-chloranil), 2,6-Dichloro-1,4-benzoquinone (2,5-DCBQ),

(2,6-DCBQ),

2-chloro-1,4-benzoquinone

methochloride (2-PAM) were purchased

2,5-dichloro-1,4-benzoquinone (2-CBQ)

from

Pyridine-2-aldoxime

Sigma-Aldrich.

HPLC-grade

acetonitrile was obtained from J. & K. Chemical Ltd. 2,3-dichloro-1,4-benzoquinone (2,3-DCBQ),

2,3,5-trichloro-1,4-benzoquinone

(TrCBQ),

2-Cyano-1-methylpyridinium iodine (2-CMP), N-methyl-2-Pyridinecarboxaldehyde chloride (N-Me-2-PCA),

15

N-labeled 2-PAM ([15N]-2-PAM) were synthesized

according to the literature methods. DMPO (5,5-dimethyl-1-pyrroline-N-oxide) and BMPO (5-tert-Butoxycarbonyl-5-methyl-1-pyrroline-N-oxide) were purchased from DOJINDO MOLECULAR TECHNOLOGIES, INC. The chemicals were used as received without further purification.

ESR studies: Reactions were carried out in Chelex-treated phosphate buffer (100 mM, pH 7.4). The solutions were recorded at 1 min after the interaction between 2-PAM and TCBQ at room temperature under normal room-lighting conditions on a Bruker ESR 300 spectrometer operating at 9.86 GHz and a cavity equipped with a Bruker Aquax liquid sample cell. Typical spectrometer parameters were as follows: scan range, 100 G; center field, 3511 G; time constant, 327.68 ms; sweep time, 61.44 s; modulation amplitude, 1.00 G; modulation frequency, 100 kHz; receiver gain, 2.52 × 103; and microwave power, 20 mW. The hyperfine splitting constants were measured by using the simulation software WinSim (version 0.96) (NIEHS). The experiments were repeated twice, and the SD was less than 10%.

Quantificaton of N-Me-2-PCA, 2-CMP and 2-PAM: The

interactions

between

TCBQ

and

2-PAM

were

monitored

by

a

TSQ-ESI-HPLC-MS (Thermo Scientific, TSQ Quantum Access), and 4.6 × 250 mm 5 μm Amide Analytical column (SUPELCO Analytical), using Chelex-treated CH3COONH4 buffer (0.01 M, pH 7.4) at room temperature. The mobile phase was 3/1000 acetic acid: acetonitrile (90:10), and the flow rate was 1.0 mL/min. The mass spectrometer was operated in the positive-ion mode. We used Selective Reaction Monitoring (SRM), N-Me-2-PCA, 2-CMP and 2-PAM were monitored by the transition ion of m/z 122.06→m/z 94.07, m/z 119.06→m/z 78.03, m/z 137.07→m/z 93.06 in the reaction mixtures in HPLC-MS/MS analysis, and the collision energy was 18, 23 and 20eV, respectively. The data were collected by Thermo Xcalibur Data Acquisition Workstation. The fragmentor voltage was 90 V, nitrogen was used as nebulizer gas, and the desolvation gas (nitrogen) was heated to 300 ºC and delivered at a flow rate of 9.0 L/min. The capillary voltage was set at 3500 V. The injection volume was 10 µL of the reaction mixture. The experiments were repeated twice, and the SD was less than 5%.

Synthesis of N-methyl-2-Pyridinecarboxaldehyde chloride (N-Me-2-PCA): 2-Pyridinecarboxaldehyde (10.5 g) was first dissolved in acetonitrile (40 mL). Iodomethane (9.5 mL) was added to this solution slowly with stirring and the solution was refluxed overnight. Yellow solid N-methyl-2-Pyridinecarboxaldehyde chloride (N-Me-2-PCA) was collected by vacuum filtration, and washed by ethyl alcohol. The yellow solid was taken to dryness under vacuum overnight. The 1H NMR and

13

C

NMR spectra were recorded on a Bruker ARX 400 spectrometer using D2O as solvent. 1

H NMR δ = 4.20 (s, 3H), 6.423 (s, 1H), 8.003 (t, 1H), 8.358 (d, 1H), 8.587 (t, 1H),

8.790 (s, 1H); 13C NMR δ = 45.81, 85.25, 125.35, 127.65, 146.66, 147.22, 155.14.

Synthesis of 15N-labled 2-PAM ([15N]-2-PAM): Pyridine-2-carbaldehyde (1.071 g) was dissolved in water (15 mL), and [15N]-NH3OH·HCl (0.742 g) was added. The solution was neutralized with aqueous

sodium hydroxide and the yellow oil which separated was extracted into ether. The ether extract was dried and the ether removed to give a pale yellow viscous oil. Then dissolved in ethyl alcohol (8 mL), iodomethane (1.9 mL) was added to this solution slowly with stirring and the solution was refluxed overnight. Yellow solid 15N-labeled 2-PAM was collected by vacuum filtration, and washed by ethyl acetate. The yellow solid was taken to dryness under vacuum overnight.

Statistics: The SD value was calculated by Excel 2016 function STDEV.

Acknowledgments This work was supported by Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDB01020300), and the National Natural Science Foundation of China (21377158, 21577149, 21477139, 21237005, 21621064, 21777180 and 21321004). References [1]. [2].

[3].

[4]. [5].

[6]. [7]. [8].

[9].

Bolton JL, Trush MA, Penning TM, Dryhurst G, & Monks TJ Role of quinones in toxicology. Chem. Res. Toxicol. 13 (2000) 135-160. Song Y, Wagner BA, Witmer JR, Lehmler H-J, & Buettner GR Nonenzymatic displacement of chlorine and formation of free radicals upon the reaction of glutathione with PCB quinones. Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 9725-9730. Zhu B-Z & Shan G-Q Potential mechanism for pentachlorophenol-induced carcinogenicity: a novel mechanism for metal-independent production of hydroxyl radicals. Chem. Res. Toxicol. 22 (2009) 969-977. Guyton KZ, et al. Carcinogenicity of pentachlorophenol and some related compounds. The Lancet Oncology 17 (2016) 1637-1638. Sorokin A, Meunier B, & Séris J-L Efficient oxidative dechlorination and aromatic ring cleavage of chlorinated phenols catalyzed by iron sulfophthalocyanine. Science 268 (1995) 1163-1166. Meunier B Catalytic degradation of chlorinated phenols. Science 296 (2002) 270-271. Gupta SS, et al. Rapid total destruction of chlorophenols by activated hydrogen peroxide. Science 296 (2002) 326-328. Zhao Y, Qin F, Boyd JM, Anichina J, & Li X-F Characterization and determination of chloro- and bromo-benzoquinones as new chlorination disinfection byproducts in drinking water. Anal. Chem. 82 (2010) 4599-4605. Huang R, et al. Ultra pressure liquid chromatography–negative electrospray ionization mass spectrometry determination of twelve halobenzoquinones at ng/L

[10].

[11].

[12]. [13].

[14]. [15]. [16].

[17].

[18].

[19].

[20].

[21].

[22]. [23].

[24].

[25].

Levels in drinking water. Anal. Chem. 85 (2013) 4520-4529. Li J, Wang W, Moe B, Wang H, & Li X-F Chemical and toxicological characterization of halobenzoquinones, an emerging class of disinfection byproducts. Chem. Res. Toxicol. 28 (2015) 306-318. Kiderlen D, Eyer P, & Worek F Formation and disposition of diethylphosphoryl-obidoxime, a potent anticholinesterase that is hydrolyzed by human paraoxonase (PON1). Biochem. Pharmacol. 69 (2005) 1853-1867. Eyer PA & Worek F Oximes. Chemical Warfare Agents: Toxicology and Treatment eds Marrs TC, Maynard RL, & Sidell FR (Wiley), second Ed, (2007) pp 305-329. Milan Jokanovic MP Pyridinium oximes as cholinesterase reactivators. structure-activity relationship and efficacy in the treatment of poisoning with organophosphorus compounds. Curr. Med. Chem. 16 (2009) 2177-2188. Kalisiak J, Ralph EC, Zhang J, & Cashman JR Amidine-oximes: reactivators for organophosphate exposure. J. Med. Chem. 54 (2011) 3319-3330. Mercey G, et al. Reactivators of acetylcholinesterase inhibited by organophosphorus nerve agents. Acc. Chem. Res. 45 (2012) 756-766. Lee JY, Lee YH, & Byun YG Detection of chemical warfare nerve agents via a Beckmann fragmentation of aldoxime. Phosphorus, Sulfur Silicon Relat. Elem. 187 (2012) 641-649. Sharma R, et al. Synthesis and in-vitro reactivation screening of imidazolium aldoximes as reactivators of sarin and VX-inhibited human acetylcholinesterase (hAChE). Chem. Biol. Interact. 259 (2016) 85-92. Wei Z, et al. Conjugates of salicylaldoximes and peripheral site ligands: Novel efficient nonquaternary reactivators for nerve agent-inhibited acetylcholinesterase. Biorg. Med. Chem. 25 (2017) 4497-4505. Zhu B-Z, Har-El R, Kitrossky N, & Chevion M New modes of action of desferrioxamine: scavenging of semiquinone radical and stimulation of hydrolysis of tetrachlorohydroquinone. Free Radic. Biol. Med. 24 (1998) 360-369. Zhu B-Z, Zhu J-G, Mao L, Kalyanaraman B, & Shan G-Q Detoxifying carcinogenic polyhalogenated quinones by hydroxamic acids via an unusual double Lossen rearrangement mechanism. Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 20686-20690. Shan G-Q, et al. A combined experimental and computational investigation on the unusual molecular mechanism of the Lossen rearrangement reaction activated by carcinogenic halogenated quinones. J. Org. Chem. 80 (2015) 180-189. Xie L-N, et al. Unusual Double Beckmann Fragmentation Reaction under Physiological Conditions. J. Org. Chem. 82 (2017) 13084-13092. Zhu BZ, Kalyanaraman B, & Jiang GB Molecular mechanism for metal-independent production of hydroxyl radicals by hydrogen peroxide and halogenated quinones. Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 17575-17578. Zhu B-Z, Zhao H-T, Kalyanaraman B, & Frei B Metal-independent production of hydroxyl radicals by halogenated quinones and hydrogen peroxide: an ESR spin trapping study. Free Radic. Biol. Med. 32 (2002) 465-473. Zhu BZ, et al. Mechanism of metal-independent decomposition of organic hydroperoxides and formation of alkoxyl radicals by halogenated quinones. Proc.

[26].

[27].

[28].

[29].

[30].

[31].

[32].

[33].

[34].

[35]. [36].

[37].

[38].

[39].

Natl. Acad. Sci. U. S. A. 104 (2007) 3698-3702. Zhu BZ, et al. Metal-independent decomposition of hydroperoxides by halogenated quinones: detection and identification of a quinone ketoxy radical. Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 11466-11471. Qin H, et al. Molecular mechanism of metal-independent decomposition of lipid hydroperoxide 13-HPODE by halogenated quinoid carcinogens. Free Radic. Biol. Med. 63 (2013) 459-466. Huang CH, et al. The first purification and unequivocal characterization of the radical form of the carbon-centered quinone ketoxy radical adduct. Chem. Commun. (Camb.) 49 (2013) 6436-6438. Huang C-H, et al. Molecular Mechanism of Metal-Independent Decomposition of Organic Hydroperoxides by Halogenated Quinoid Carcinogens and the Potential Biological Implications. Chem. Res. Toxicol. 28 (2015) 831-837. Shao J, Huang C-H, Kalyanaraman B, & Zhu B-Z Potent methyl oxidation of 5-methyl-2 ′ -deoxycytidine by halogenated quinoid carcinogens and hydrogen peroxide via a metal-independent mechanism. Free Radic. Biol. Med. 60 (2013) 177-182. Jia S, Zhu B-Z, & Guo L-H Detection and mechanistic investigation of halogenated benzoquinone induced DNA damage by photoelectrochemical DNA sensor. Anal. Bioanal. Chem. 397 (2010) 2395-2400. Chen H-M, Zhu B-Z, Chen R-J, Wang B, Jr., & Wang Y-J The Pentachlorophenol Metabolite Tetrachlorohydroquinone Induces Massive ROS and Prolonged p-ERK Expression in Splenocytes, Leading to Inhibition of Apoptosis and Necrotic Cell Death. PLoS One 9 (2014) e89483. Zhao B-L, et al. Redox-active quinones induces genome-wide DNA methylation changes by an iron-mediated and Tet-dependent mechanism. Nucleic Acids Res. 42 (2014) 1593-1605. Qu N, et al. The Unexpected and Exceptionally Facile Chemical Modification of the Phenolic Hydroxyl Group of Tyrosine by Polyhalogenated Quinones under Physiological Conditions. Chem. Res. Toxicol. 29 (2016) 1699-1705. Madden KP & Taniguchi H The role of the DMPO-hydrated electron spin adduct in DMPO-OH spin trapping. Free Radic. Biol. Med. 30 (2001) 1374-1380. Bauer NA, Hoque E, Wolf M, Kleigrewe K, & Hofmann T Detection of the formyl radical by EPR spin-trapping and mass spectrometry. Free Radic. Biol. Med. 116 (2018) 129-133. Hawkins CL & Davies MJ The role of reactive N-bromo species and radical intermediates in hypobromous acid-induced protein oxidation. Free Radic. Biol. Med. 39 (2005) 900-912. Bhattacharjee S, et al. Site-specific radical formation in DNA induced by Cu(II)– H2O2 oxidizing system, using ESR, immuno-spin trapping, LC-MS, and MS/MS. Free Radic. Biol. Med. 50 (2011) 1536-1545. Walton JC The Oxime Portmanteau Motif: Released Heteroradicals Undergo Incisive EPR Interrogation and Deliver Diverse Heterocycles. Acc. Chem. Res. 47 (2014) 1406-1416.

[40]. [41].

[42].

[43]. [44]. [45].

[46].

[47].

[48]. [49].

[50].

Li M, Chen Q, Ma T, & Yu X Targeting reactive nitrogen species suppresses hereditary pancreatic cancer. Proc. Natl. Acad. Sci. U. S. A. 114 (2017) 7106-7111. Wiseman H & Halliwell B Damage to DNA by reactive oxygen and nitrogen species: role in inflammatory disease and progression to cancer. Biochem. J. 313 (1996) 17-29. Griller D, Mendenhall G, Van Hoof W, & Ingold K Kinetic applications of electron paramagnetic resonance spectroscopy. XV. Iminyl radicals. J. Am. Chem. Soc. 96 (1974) 6068-6070. Zard SZ Iminyl Radicals: A Fresh Look at a Forgotten Species (and Some of its Relatives). Synlett 1996 (1996) 1148-1154. McCarroll AJ & Walton JC Enhanced radical delivery from aldoxime esters for EPR and ring closure applications. Chem. Commun. (2000) 351-352. Bencivenni G, et al. Iminyl Radicals from α-Azido o-Iodoanilides via 1,5-H Transfer Reactions of Aryl Radicals: New Transformation of α-Azido Acids to Decarboxylated Nitriles. J. Org. Chem. 73 (2008) 4721-4724. McBurney RT, Slawin AM, Smart LA, Yu Y, & Walton JC UV promoted phenanthridine syntheses from oxime carbonate derived iminyl radicals. Chem. Commun. 47 (2011) 7974-7976. Mahajan PS, Humne VT, Tanpure SD, & Mhaske SB Radical Beckmann Rearrangement and Its Application in the Formal Total Synthesis of Antimalarial Natural Product Isocryptolepine via C–H Activation. Org. Lett. 18 (2016) 3450-3453. Dawson RM Review of oximes available for treatment of nerve agent poisoning. J. Appl. Toxicol. 14 (1994) 317-331. Medicis JJ, Stork CM, Ann Howland M, Hoffman RS, & Goldfrank LR Pharmacokinetics Following a Loading Plus a Continuous Infusion of Pralidoxime Compared with the Traditional Short Infusion Regimen in Human Volunteers. J. Toxicol. Clin. Toxicol. 34 (1996) 289-295. Kassa J Review of Oximes in the Antidotal Treatment of Poisoning by Organophosphorus Nerve Agents. J. Toxicol. Clin. Toxicol. 40 (2002) 803-816.

Highlights ► TCBQ activated 2-PAM to generate reactive iminyl radicals in two-consecutive steps. ► Analogous radical homolysis mechanism was observed with other haloquinones & aldoximes. ► This is the 1st detection of iminyl radical under normal physiological condition. ► The finding can explain partial protection by 2-PAM against TCBQ-induced damages. ► It may also explain some of the side-toxic effects induced by pyridinium aldoximes.

O Cl

Nucleophilic Substitution

Cl

Cl

Cl

O

N

N H

O

Cl

O Cl

Cl

Cl

O O

N H

I

2-PAM

TCBQ

N

1st Radical Homolysis 2nd Radical Homolysis

N

N

H N

O

O N

Cl

O Cl

H II

O O

2-PAM

Cl

Cl

Cl

O

N O

TrCBQ-O

N H