Development and Application of a Peroxyl Radical Clock Approach for Measuring Both Hydrogen-Atom Transfer and Peroxyl Radical Addition Rate Constants

pubs.acs.org/joc

Article

Development and Application of a Peroxyl Radical Clock Approach

for Measuring Both Hydrogen-Atom Transfer and Peroxyl Radical

Addition Rate Constants

Quynh Do, David D. Lee, Andrew N. Dinh, Ryan P. Seguin, Rutan Zhang, and Libin Xu*

Cite This: https://dx.doi.org/10.1021/acs.joc.0c01920

Read Online

sı

*

Metrics & More

Article Recommendations

Supporting Information

ACCESS

ABSTRACT: The rate-determining step in free radical lipid

peroxidation is the propagation of the peroxyl radical, where

generally two types of reactions occur: (a) hydrogen-atom transfer

(HAT) from a donor to the peroxyl radical; (b) peroxyl radical

addition (PRA) to a “CC” double bond. Peroxyl radical clocks

have been used to determine the rate constants of HAT reactions

(k_H), but no radical clock is available to measure the rate constants

of PRA reactions (k_add). In this work, we modiﬁed the analytical

approach on the linoleate-based peroxyl radical clock to enable the

simultaneous measurement of both k_Hand k_add. Compared to the

original approach, this new approach involves the use of a strong

reducing agent, LiAlH₄, to completely reduce both HAT and PRA-

derived products and the relative quantitation of total linoleate oxidation products with or without reduction. The new approach was

then applied to measuring the k_Hand k_addvalues for several series of organic substrates, including para- and meta-substituted

styrenes, substituted conjugated dienes, and cyclic alkenes. Furthermore, the k_Hand k_addvalues for a variety of biologically important

lipids were determined for the ﬁrst time, including conjugated fatty acids, sterols, coenzyme Q10, and lipophilic vitamins, such as

vitamins D₃and A.

nation (Figure 1).^2,4The rate-limiting step in this mechanism

is the propagation step in which the lipid peroxyl radical can

undergo two types of reactions: hydrogen (H)-atom transfer

(HAT) and peroxyl radical addition (PRA). In the HAT

reaction, the peroxyl radical abstracts a H atom from another

lipid or oxidizable substrate, generating a lipid hydroperoxide,

INTRODUCTION

■

Despite the crucial role of molecular oxygen in cell metabolism

and energy production, imbalance between pro-oxidant and

antioxidant cellular processes can lead to an increased level of

reactive oxygen species, a state generally called “oxidative

stress”. Oxidative stress can result in DNA damage and alter

while in the PRA reaction, the peroxyl radical adds to a “C

C” double bond. This PRA product can then either undergo

intramolecular homolytic substitution (S_Hi) to form an epoxide

or react with another molecular oxygen to form a new peroxyl

radical. Both the HAT and PRA reactions generate new

radicals that can continue the free radical chain oxidation by

reacting with another lipid molecule or oxidizable substrate.

The direct measurement of HAT rate constants, k_H, for a

variety of organic substrates requires the knowledge of the

rates of polymerization (R_p), initiation (R_i), and termination

(R_t); however, R_tcan only be determined using the rotating-

sector method.²⁰⁻²²On the other hand, a radical clock, which

cell proliferation, diﬀerentiation, or metabolism.¹Lipids are

among the primary targets of attack by reactive oxygen species,

leading to subsequent free radical chain reactions with

molecular oxygen, termed lipid peroxidation. Since lipids are

essential components of the cell membrane, lipid peroxidation

is associated with multiple human conditions and diseases,²⁻⁴

including atherosclerosis,⁵diabetes,⁶cancer,^7,8neurodegener-

ative disorders,⁹⁻¹¹and certain cholesterol biosynthesis

disorders.¹²More recently, lipid peroxidation was discovered

to be associated with ferroptosis, a regulated form of cell death

that is morphologically, biochemically, and genetically distinct

from other types of cell death such as apoptosis, necrosis, and

autophagy.^13,14More importantly, nonenzymatic lipid perox-

idation, or autoxidation, was found to be the key eﬀector in the

induction of ferroptosis.¹⁵⁻¹⁷

Received: August 8, 2020

The lipid autoxidation mechanism, kinetics, and product

distribution have been studied intensively for decades.^2,4,18,19

The autoxidation process occurs via a three-step mechanism

including the following: initiation, propagation, and termi-

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

A

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Figure 1. Sequence of the free radical chain oxidation of lipids.

Figure 2. Original linoleate peroxyl radical clock.²³

Figure 3. Linoleate peroxyl radical clock in the presence of both HAT and PRA reactions.

utilizes a known rate constant of a unimolecular reaction to

measure an unknown rate constant of a competing bimolecular

reaction, is a more convenient method to indirectly measure

k_H.²³⁻²⁶Speciﬁcally, a linoleate-based peroxyl radical clock,

which utilizes the competition between the β-fragmentation

reactions with known rate constants and the bimolecular HAT

reaction, was developed to measure k_Hfor a variety of lipids,

including polyunsaturated fatty acids (PUFAs) and sterols

(Figure 2).²³⁻²⁶Particularly, the reaction between linoleate

and molecular oxygen can generate peroxyl radicals at either

the 9- or 13-position of the carbon chain. These 9- or 13-

peroxyl radicals can then undergo HAT to yield either 9-trans,

B

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Figure 4. Series of substrates that are examined in this study: (a) styrene and its para-substituted derivatives, (b) meta-substituted styrene

derivatives, (c) styrenes with methyl substituent at the α- and β-position, (d) conjugated dienes with diﬀerent methyl substituents, (e) cyclic

hydrocarbons, and (f) biologically important lipids.

cis (t,c)- or 13-t, c-hydroperoxide products, which are termed

HPODEs, or they can also undergo the competing β-

fragmentation reaction, leading to 13-trans, trans (t,t)- and 9-

t,t-HPODEs. Therefore, linoleate peroxyl radicals, with known

rate constants for β-fragmentation, can be used to “clock” the

unknown rate constants of HAT reactions via eq 1:

While the HAT reaction mechanism and rate constants have

been studied extensively via the radical clock method, there has

not been any method established to measure the rate constants

for the PRA reaction, k_add. Epoxides, one of the addition

products, were identiﬁed among the oxidation products from

sterols, such as cholesterol and 7-DHC.^12,29,30In addition,

intramolecular PRA reactions were discovered to occur in

cardiolipins with multiple linoleate chains, once one chain is

oxidized to linoleate peroxyl radical.³¹This is evidence that

PRA also contributes signiﬁcantly to the propagation step, and

thus, its rate constants should also be considered in the study

of lipid peroxidation kinetics. This is a major gap in the ﬁeld

that this study aims to ﬁll. The sum of k_Hand k_add, also deﬁned

as the total propagation rate constant (k_p) of the lipid

peroxidation of a substrate, can also be measured with the

inhibited autoxidation method using a set of calibrated

reference inhibitors.³²However, this method cannot diﬀer-

entiate between the two rate constants either. To fully

understand the complex nature of the lipid peroxidation

mechanism, kinetics, and products, we modiﬁed the current

[t, c‐HPODEs]

[t, t‐HPODEs]

[9‐t, c‐HPODEs + 13‐t, c‐HPODEs]

[9‐t, t‐HPODEs + 13‐t, t‐HPODEs]

=

k_H[Substrate]

+ 0.16

(1)

214

Here, 214 and 0.16 are β-fragmentation-dependent con-

stants.²³Based on the measurements using the peroxyl radical

clock method, cholesterol was found to be a moderately

oxidizable lipid with a k_Hof 11 M⁻¹s⁻¹at 37 °C, but 7-

dehydrocholesterol (7-DHC), a biosynthetic precursor of

cholesterol and vitamin D₃,^27,28was found to be the most

reactive lipid molecule so far, with a k_Hof 2260 M⁻¹s⁻¹.²⁴

C

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Figure 5. Proposed new radical clock approach and the structures of 13-t,c-, 9-t,c-, 13-t,t-, and 9-t,t-HODOL. (HPODE: hydro-

peroxyoctadecadienoic acid, i.e., nonreduced products from HAT reaction; HODOL: hydroxyoctadecadienol, i.e., reduced products from both

HAT and PRA reactions; t,c: trans, cis; t,t: trans, trans; k_H: HAT rate constant; k_add: PRA rate constant.)

linoleate peroxyl radical clock to account for the contribution

of the PRA reaction (Figure 3).

not strong enough to reduce all the linoleate-derived products,

particularly those from the PRA reactions, which are

dialkylperoxides and sometimes in the form of peroxide

polymers.²Attempts using sodium borohydride under various

conditions were also unsuccessful, leading to the ratios

between the concentration of reduced and nonreduced

oxidation products that are less than one. We found that the

complete reduction of all oxidation products could only be

achieved using the strong reducing agent lithium aluminum

hydride (LiAlH₄). It should be noted here that LiAlH₄reduces

both the peroxyl bonds and the ester moiety in MeLn to

alcohols. Importantly, these diols are a completely new set of

products from the oxidation of linoleate that have never been

characterized before as readouts for radical clocks. We termed

these linoleate-derived alcohols, or hydroxyoctadecadienols,

HODOLs (structures shown in Figure 5).

We report here the development of a new analytical

approach based on the original linoleate-based peroxyl radical

clock, which enables the simultaneous measurement of the rate

constants of both HAT and PRA reactions. We ﬁrst explain the

design of the new peroxyl radical clock approach. We then

apply this new approach to elucidate the values of k_Hand k_add

for a variety of substrates, including para- and meta-substituted

styrenes, double bond-substituted styrenes, conjugated dienes

with methyl substituent at diﬀerent positions, cyclic com-

pounds, and biologically important lipids (Figure 4). Based on

these results, we will discuss the factors that aﬀect the

contribution of each reaction to the overall propagation step.

In the end, we will discuss the limitations and the biological

implications of the new approach.

Our new peroxyl radical clock approach can be summarized

into two main steps: to measure the sum of k_Hand k_addand to

determine the ratio of k_add/k_H(Figure 5). The sum of k_Hand

RESULTS

■

Design of the New Peroxyl Radical Clock. Our new

approach was developed based on the autoxidation of methyl

linoleate (MeLn) as its mechanism is well studied. In addition,

our preliminary studies had shown that the high-performance

liquid chromatography (HPLC) chromatograms of MeLn-

based oxidation products had better separation compared to

that of linoleic acid, which enables better product analysis. To

elucidate the values of k_Hand k_add, it is crucial to capture all the

linoleate-derived peroxide products by fully reducing the

hydrogen peroxyl group in HPODEs from the HAT reaction

and the alkyl peroxyl group in addition products from the PRA

reaction. We used the co-oxidation reaction of MeLn and

styrene in benzene to examine diﬀerent reducing agents and

conditions because styrene can only undergo PRA reactions

(see the Experimental Section). After the peroxidation

reaction, the reaction mixture was split into two halves.

While the ﬁrst half was analyzed directly, the second half was

reduced prior to analysis. This ratio between the total reduced

oxidation products and the total nonreduced oxidation

products is expected to be equal to or larger than one if all

the oxidation products are fully reduced. Even though

triphenylphosphine was commonly used in the classic radical

clock approach as a reducing agent,^23,24we found that it was

k

_add, which is equal to k_p, can be elucidated via plotting the

ratio [t,c-HODOLs]/[t,t-HODOLs] at diﬀerent substrate

concentrations. Speciﬁcally, when an oxidizable substrate can

undergo both HAT and PRA reactions, the ratio [t,c-

HODOLs]/[t,t-HODOLs] will be mediated by the competi-

tion from the HAT, β-fragmentation, and PRA reactions.

Therefore, k_pcan now be calculated from eq 2:

k_p[Substrate]

[t, c‐HODOLs]

[t, t‐HODOLs]

=

+ 0.16

214

(2)

On the other hand, the ratio k_add/k_Hcould be elucidated by

plotting the ratio between the concentration of the total

HODOLs, which are reduced oxidation products derived from

both the HAT and PRA reactions, termed HODOL_total, and of

the total HPODEs, which are nonreduced oxidation products

derived from only the HAT reaction, termed HPODE_total, at

diﬀerent substrate concentrations. Thus, eq 3 describes the

ratio [HODOL_total]/[HPODE_total] as a function of substrate

concentration.

D

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Figure 6. Determination of k_Hand k_addfor a substrate that can undergo (a) both reactions, (b) HAT reaction only, and (c) PRA reaction only.

Here, the k_pvalue is calculated using the slope of the [Substrate] vs [t,c-HODOLs]/[t,t-HODOLs]. The ratio k_add/k_His generated from the ﬁtting

of the [Substrate] vs [HODOL_total]/[HPODE_total] plot using Prism. The measurement for each substrate was performed in triplicate.

Figure 7. HPLC chromatogram at 234 nm for the (a) HPODE and (b) HODOL products eluting with solvent hexane:isopropyl alcohol = 99.5:0.5

and 95.5:4.5, respectively. The internal standard peak was monitored at 254 nm.

[HODOL_total

[HPODE_total

]

DOL_total]/[HPODE_total] plots for 1,3-cyclohexadiene (1,3-

CHD), a substrate that can undergoes both reactions, is

shown in Figure 6a. When a substrate can only undergo the

HAT reaction, such as 1,4-CHD, eq 3 would be rewritten as eq

4 and the linear ﬁt of the [Substrate] vs [HODOL_total]/

[HPODE_total] data would give a slope close to 0 (Figure 6b).

In this case, the value of k_Hwould be reported as the value of

k_p. In contrast, when PRA is predominant compared to HAT,

such as in the oxidation of styrene (Sty), eq 3 would become

eq 5, and the linear ﬁt of [Substrate] vs [HODOL_total]/

[HPODE_total] data would give a nonzero slope (Figure 6c).

Here, the value of k_addwould be equal to the value of k_p.

k_MeLn[MeLn] + k_H[Substrate] + k_add[Substrate]

k_MeLn[MeLn] + k_H[Substrate]

=

k_add[Substrate]

k_MeLn[MeLn] + k_H[Substrate]

=

+ 1

k_add

k_H

k_H[Substrate]

=

+ 1

k_MeLn[MeLn] + k_H[Substrate]

k_add

k_H[Substrate]

k_H

[HODOL_total

]

+ 1

k_Hdominant:

= 1

18.7 + k_H[Substrate]

(3)

[HPODE_total

]

(4)

Here, k_MeLnrepresents the total propagation rate constant of

MeLn, which is 62 M⁻¹s⁻¹as measured by the rotating-sector

method.²²In addition, [MeLn], which is ﬁxed at 0.3 M, is the

concentration of MeLn used to set up the radical clock

experiment. Therefore, the term k_MeLn[MeLn] is a constant

that is equal to 18.7 s⁻¹. By ﬁtting the [Substrate] vs

[HODOL_total]/[HPODE_total] curve to eq 3 using Prism

software, we could obtain the value of k_add/k_H. Now, the

individual value for each rate constant could be solved from

the sum and the ratio of k_Hand k_add. An example on the

determination of k_p, k_add, and k_Hvalues from the [Substrate] vs

[t,c-HODOLs]/[t,t-HODOLs] and [Substrate] vs [HO-

[HODOL_total

]

k_add[Substrate]

k

_adddominant:

=

+ 1

[HPODE_total

]

18.7

(5)

For this new peroxyl radical clock approach, normal phase

HPLC/UV was utilized to quantify the relative amounts of

HPODE and HODOL products, respectively, to an internal

standard, 9-anthracenemethanol (Figure 7) (see the Exper-

imental Section).

Application of the New Peroxyl Radical Clock

Approach to Examining the Reactivities of Substrates

That Can Only Undergo the PRA Mechanism. We ﬁrst

E

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

applied the new approach to measure the rate constants of

para- and meta-substituted styrenes (Figure 4a and b). Since

styrene and its para- and meta-substituted derivatives can only

undergo PRA in reactions with MeLn, this series is useful not

only in assessing the application of the new approach but also

in determining the substituent eﬀects on the reactivity of the

PRA mechanism. The [Substrate] vs [HODOL_total]/[HPO-

DE_total] curve for styrene and its derivatives conﬁrmed that

they are indeed k_adddominant (Figure S1 in the Supporting

Information). The value of k_add, which is equal to k_p, for each

compound was determined and is reported in Table 1. We

found that the experimental k_addvalue of styrene is on the same

order of magnitude as the value measured previously using the

rotating-sector method at 30 °C.³³

provide further insights into our results, we calculated the

activation energies for the addition reaction between the

peroxyl radical and each of the para-substituted styrenes using

the B3LYP/6-31++G(d,p) method (Table 1). In the computa-

tional model of the reaction transition state, we used the

methyl peroxyl radical as a simpliﬁed analogue for the MeLn

peroxyl radical to expedite the calculations. Consistent with the

measured k_addvalues, we observed a decrease in the computed

energy barriers for Sty, 4-MeSty, and 4-MeOSty, respectively.

There was also no signiﬁcant diﬀerence in the activation

energies of 4-ClSty, 4-FSty, and 4-NO₂Sty from that of Sty,

which agreed with our experimental data. The experimental

k

_addvalues also appear to align better with the trend predicted

by the computed spin density than the atomic polar tensor

charge, which represents the partial charge distribution (Figure

9a and 9b).

Table 1. PRA Rate Constants (k_add) and Activation Energies

We then attempted to generate a Hammett correlation by

plotting log(k_X/k_H) versus either the σ_para-scale³⁶or the radical

σ-scale (σ_rad)³⁷(Figure S16a and b). However, both Hammett

parameters did not allow us to derive a good linear correlation

between substituent eﬀects and the rate constant of the PRA

reaction. Therefore, we utilized the extended Hammett

equation:

(E_a) of Para- and Meta-Substituted Styrene Derivatives at 37

°C

substrate

k_add(M⁻¹s⁻¹

)

E_a(kcal/mol)

lit. k_add(M⁻¹s⁻¹

)

a

Sty

17

24

56

17

13

23

27

20

15

1

3

2

1

19.5

19.2

18.8

19.4

19.5

19.4

19.7

19.9

19.7

41

4-MeSty

4-MeOSty

4-ClSty

4-FSty

4-NO₂Sty

3-MeSty

3-MeOSty

3-ClSty

3-FSty

na

log k_X/k_H= ρσ⁺+ σ

(6)

rad

to derive the correlation between the substituent eﬀects and

the computed activation energies (Figure 9e).³⁸Note that

since the reaction rates appear to be mediated primarily

through radical stabilization eﬀects and the transition state

contains a partial positive charge, the use of the σ⁺-scale³⁹is

more appropriate than the σ-scale.³⁶The values of log(k_X/k_H),

σ⁺, and σ_radfor each substituent were ﬁt to eq 6, giving a ρ-

value of −0.66. Fitting of the calculated activation energy and

the combined Hammett parameters (σ_rad− 0.66σ⁺) with linear

regression gave eq 7 (R²= 0.85):

a

From ref 33 at 30 °C. na, not available.

Since the addition of peroxyl radical to an alkene is an

electrophilic reaction,³⁴the polar transition state of the PRA

reaction would be opposite to the carbon-radical addition to

alkenes, which was a nucleophilic process (Figure 8).³⁵

E_a= −1.04(σ − 0.66σ⁺) + 19.4 (kcal/mol)

(7)

rad

which describes the dual radical-cationic properties of the

benzylic carbon in styrene and its derivatives (Figure 9c). The

negative slope of the line also conﬁrmed that there is a partial

positive charge buildup on the benzylic carbon atom during

the transition state and that the addition of peroxyl radical to

alkene is indeed an electrophilic reaction. Fitting of the

calculated activation energy and log(k_X/k_H) also gave a good

linear correlation (R²= 0.93) with a negative slope of −1.21,

which shows a good agreement between our theoretical and

experimental data (Figure 9d). We then examined the meta-

substituted styrene derivatives to investigate how the addition

rate constants and activation energies would change with only

contribution from the polar eﬀects (Table 1). The small

changes in both rate constants and activation energies of the

meta-substituted styrene derivatives further suggested that the

contribution of polar eﬀects to the reactivity of the PRA

reaction is small. In addition, the rate constants of the meta-

substituted styrene derivatives and σ_metaparameters did not

give a good linear correlation in the Hammett plot (Figure

S16c).

Figure 8. Comparison of the polar transition state between (a) carbon

radical and (b) peroxyl radical to an alkene.

Therefore, we hypothesized that the addition rate constant

of the peroxyl radical to styrene would increase as the electron

density at the double bond increases. We indeed observed an

increase in k_addvalues of styrene derivatives with strong

electron-donating groups at the para-position such as 4-MeSty

and 4-MeOSty. However, the addition rate constants of

styrene derivatives with electron-withdrawing groups such as 4-

ClSty, 4-FSty, and 4-NO₂Sty were found to have k_addvalues

that are similar to that of styrene. This observation suggests

that radical stabilization eﬀects are more important than polar

eﬀects in inﬂuencing the reactivities of PRA to styrene. To

Application of the New Peroxyl Radical Clock

Approach to Structure−Reactivity Relationship Studies

of Diﬀerent Oxidizable Substrates. Here, we applied the

new peroxyl radical clock approach to measure the values of k_p,

k_H, and k_addfor diﬀerent series of oxidizable substrates

F

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Figure 9. (a) Spin density and (b) atomic polar tensor charge distribution in the transition states; (c) activation energies versus the combination of

radical and cation Hammett parameters; (d) activation energies versus log(k_X/k_H); (e) log(k_X/k_H) versus the combination of radical and cation

Hammett parameters for PRA reactions to the para-substituted styrene derivatives. The atoms displayed are carbon (dark gray), hydrogen (light

gray), oxygen (red), chlorine (green), ﬂuorine (light blue), and nitrogen (dark blue).

including α- and β-MeSty, conjugated dienes with diﬀerent

methyl substituents, and cyclic compounds (Figure 4c, d, and

e). The results are summarized in Table 2.

Structure−reactivity relationship studies on these substrates

allow us to examine the factors that aﬀect the contribution of

each mechanism to the overall propagation rate constant. For

example, both α- and β-MeSty have signiﬁcantly diﬀerent

reactivities toward the HAT mechanism due to the location of

their methyl substituent. Particularly, while the proposed

resulting carbon radicals formed after the abstraction of H

atom from α-MeSty span only three carbon centers, the loss of

H atom from β-MeSty potentially forms a much more

delocalized radical (Figure 10). This explains why β-MeSty

has signiﬁcant contribution from the HAT mechanism while α-

MeSty is k_adddominant. However, we are surprised by the

higher k_addvalue of β-MeSty compared to α-MeSty. The

addition of peroxyl radical into the terminal methylene group

of α-MeSty would potentially lead to the formation of a tertiary

benzylic radical, which is highly stable. While the addition of

the peroxyl radical at the same carbon of β-MeSty would

G

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Table 2. Free Radical Oxidation Propagation (k_p), Hydrogen-Atom Transfer (k_H), and Peroxyl Radical Addition (k_add) Rate

a

Constants of Diﬀerent Organic Oxidizable Substrates Measured by the New Peroxyl Radical Clock at 37 °C

substrate

k_p(M⁻¹s⁻¹

)

k_add/k_H

k_H(M⁻¹s⁻¹

)

k_add(M⁻¹s⁻¹

)

lit. k_H(M⁻¹s⁻¹

)

lit. k_add(M⁻¹s⁻¹

)

a

α-MeSty

β-MeSty

1,3-pentadiene

27

63

34

154

97

65

1

2

4

1

7

k_adddominant

3.1 0.6

k_adddominant

2.4 0.5

0.5 0.1

1.0 0.3

1.1 0.2

0.95 0.12

k_Hdominant

na

15

na

29

43

81 11

243 21

449 26

478 15

26

48

34

154

68

22

84 11

266 21

429 26

na

2

4

3

na

10

a

2

51

na

3-methyl-1,3-pentadiene

2,4-dimethyl-1,3-pentadiene

1,3-hexadiene

4

3

2,4-hexadiene

165

2,5-dimethyl-2,4-hexadiene

1,3-cyclohexadiene (1,3-CHD)

1,4-cyclohexadiene (1,4-CHD)

509 73

878 35

478 15

b

220

1480

265

b

c

d

362

b

a

1,2-dihydronaphthalene

indene

141

80

100

8

5

7

0.33 0.08

0.47 0.09

1.1 0.3

106

55

48

6

3

6

35

25

52

6

3

6

52

291

128

b

a

14

2-methylindene

α-terpinene

na

1304 11

0.28 0.03

1019 24

285 24

a

b

c

d

k_Hvalues are reported as per molecule, instead of per H atom. na, not available. From ref 22 at 30 °C. From ref 23 at 37 °C. From ref 40 at 30

°C.

methyl-1,3-pentadiene even though it has more methyl

substituents. In 3-methyl-1,3-pentadiene, the peroxyl radical

can be added to both carbon 1 and 4. However, in 2,4-

dimethyl-1,3-pentadiene, the two methyl substituents at carbon

4 sterically hinder the attack of the peroxyl radical at this

position. Comparison between 1,3-pentadiene and 1,3-

hexadiene also allowed us to conﬁrm that changes in k_H

values are dependent on the stability of the radical formed

after the loss of the H atom. Speciﬁcally, after the HAT

reaction, 1,3-hexadiene would yield a secondary radical that is

much more stable than the primary radical formed from 1,3-

pentadiene. Hence, 1,3-hexadiene has a much higher k_Hvalue

and lower k_add/k_Hratio than 1,3-pentadiene.

The diﬀerence in reactivities of 1,3-CHD and 1,4-CHD

indicates that the PRA mechanism can only signiﬁcantly

contribute to k_pwhen there is a conjugated double bond

system. Speciﬁcally, 1,4-CHD has a nonconjugated double

bond system with two bis-allylic positions, which have been

found to be highly reactive toward H atom transfer.⁴¹Thus,

1,4-CHD is k_Hdominant. In contrast, 1,3-CHD has a

conjugated double bond system, which explains why there is

signiﬁcant contribution of k_addto its total propagation rate

constant. It should be noted that the k_Hvalues measured for

1,3- and 1,4-CHD are almost identical, which is consistent with

the similar C−H bond dissociation energies (BDEs) previously

reported for the two compounds.⁴²We also want to note the

discrepancy of our experimental k_Hvalue for 1,4-CHD (478

M⁻¹s⁻¹) compared to the value previously reported (1480

M⁻¹s⁻¹).²²Similar to the cases of Sty, α-MeSty, and β-MeSty,

the diﬀerence can probably be attributed to the diﬀerent

propagating radicals between the rotating-sector method and

our new peroxyl radical clock approach. Speciﬁcally, the chain-

carrying species in the oxidation of 1,4-CHD under the

rotating-sector method are mainly H−OO· radicals, which is

sterically unhindered and highly reactive, thus explaining the

larger HAT rate constant.⁴³It should be noted that our

experimental k_Hvalue for 1,4-CHD is on the same order of

magnitude as the previous results in which the chain-carrying

species are also linoleate-based peroxyl radicals.^23,40In

addition, Howard and Ingold reported predominant contribu-

Figure 10. Proposed resulting carbon radicals formed after the HAT

reactions from (a) α-MeSty and (b) β-MeSty.

generate a stable secondary benzylic carbon radical, this

addition position has a methyl substituent. Thus, one would

expect β-MeSty to have lower k_addvalue than α-MeSty. In

addition, we want to note the diﬀerence in the order of our

experimental k_addvalues for Sty, α-MeSty, and β-MeSty from

those reported by the rotating-sector method. The discrepancy

here can be attributed to the fact that the propagation rate

constants depend on the structure of not only the substrate but

also the chain-propagating peroxyl radical.²²Speciﬁcally, in the

rotating-sector method, the resulting chain-carrying species is a

tertial peroxyl radical after the addition of peroxyl radical to α-

MeSty, which is a lot more hindered than the secondary

peroxyl radicals derived from Sty and β-MeSty.²²Hence, the

rate constant for α-MeSty measured using the rotating-sector

method is much smaller than those of Sty and β-MeSty. On the

other hand, in the peroxyl radical clock method, the chain-

carrying species are the same secondary linoleate-based peroxyl

radicals for all substrates, which means the propagation rate

constants depend mostly on the structures of the substrates.

Pairwise comparison of substrates from Table 2 also

conﬁrms the electronic eﬀects on k_add. For example, structures

that contain additional methyl substituents on the double

bond, including 3-methyl-1,3-pentadiene, 2,4-dimethyl-1,3-

pentadiene, 2,4-hexadiene, and 2,5-dimethyl-2,4-hexadiene

(Figure 4d), have much higher k_addvalues compared to 1,3-

pentadiene. Similarly, the addition of a methyl substituent on

2-methylindene also doubles the k_addvalue of this substrate

compared to indene. However, it should be noted that 2,4-

dimethyl-1,3-pentadiene has a lower k_addvalue compared to 3-

H

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Table 3. Free Radical Oxidation Propagation (k_p), Hydrogen-Atom Transfer (k_H), and Peroxyl Radical Addition Rate

Constants (k_p) of Diﬀerent Biologically Important Lipids Measured by the New Peroxyl Radical Clock Approach at 37 °C

a

substrate

k_p(M⁻¹s⁻¹

)

k_add/k_H

k_H(M⁻¹s⁻¹

)

k_add(M⁻¹s⁻¹

)

lit. k_p(M⁻¹s⁻¹

)

b

7-dehydrocholesterol (7-DHC)

cholesteryl acetate

2737 83

k_Hdominant

k_adddominant

1.4 0.4

1.1 1.3

2.4 1.7

k_Hdominant

2737 83

na

2260

31.4

c

36

2

36

2

retinal

coenzyme Q10

vitamin D₃

5656 143

695 23

1031 63

na

5656 143

695 23

601 63

61 21

876 117

na

430 63

57 21

359 117

conjugated linoleic acid 18:2 (CLA 18:2)

conjugated linolenic acid 18:3 (CLA 18:3)

nonconjugated linolenic acid 18:3 (NLA 18:3)

118

1235 60

144

8

3

144

3

a

b

c

k_Hvalues are reported as per molecule, instead of per H atom. na, not available. From ref 24 at 37 °C. From ref 29 at 37 °C.

tion of the PRA mechanism to the propagation rate constants

of 1,3-CHD, 1,2-dihydronaphthalene, and indene.^22,42How-

ever, we found that the HAT mechanism contributes more

signiﬁcantly to the propagation step of 1,2-dihydronapthalene

and indene, and both mechanisms contribute similarly to the

propagation rate constant of 1,3-CHD. This discrepancy may

be attributed to the fact that in the rotating-sector method, the

H-atom abstracting species are polymeric peroxyl radicals,

R(OOR)_nOO·, derived from the substrate.²²Furthermore, the

k_addvalues for 1,2-dihydronaphthalene and indene are on the

same order of magnitude as those of the styrene derivatives

reported here, suggesting consistency when using the radical

clock method. Because of the diﬀerent propagating radicals

involved between the rotating-sector method and the peroxyl

radical clock method, we suggest that the rate constants

measured using the same method would provide better

comparison of the reactivity of the oxidizable molecules.

Examination of Reactivities of Biologically Important

Lipids. The values of k_p, k_H, and k_addfor diﬀerent biologically

important lipids including sterols, lipophilic vitamins, and

PUFAs were also measured using the new peroxyl radical clock

and are reported in Table 3 (Figure 4f).

Due to the low solubility of cholesterol in benzene and

hexane, which caused diﬃculties in both the experimental

setup and product analyses, we opted to use its ester analogue,

cholesteryl acetate, as the substrate instead. Even though

cholesteryl acetate is more reactive toward free radical

oxidation than cholesterol, their product distribution was

reported to be the same.²⁹The k_pvalues of cholesteryl acetate

and 7-DHC measured by the new radical clock are 36 and

2737 M⁻¹s⁻¹, respectively. Both values are close to the

previously reported values of 31.4 and 2260 M⁻¹s⁻¹, which

were measured using the original peroxyl radical clock

method.^24,29These results validate the applicability of the

new approach in examining the rate constants of lipid

peroxidation. We found the HAT mechanism to be

predominant in cholesteryl acetate, which is consistent with

the previous ﬁnding that cholesterol addition products only

account for 12% of the total oxidation products.²⁹Due to the

conjugated double bonds in the structure of 7-DHC, we

expected to observe signiﬁcant contribution from the PRA

mechanism into the overall rate constant. However, surpris-

ingly, we found 7-DHC to also be k_Hdominant. In addition,

coenzyme Q10 and lipophilic vitamins such as vitamin D₃and

retinal were all found to be highly reactive toward free radical

oxidation with k_pvalues of 695, 1031, and 5656 M⁻¹s⁻¹,

respectively. Signiﬁcantly, the k_pvalue of retinal suggested that

this compound would replace 7-DHC as the most reactive

biological compound toward free radical oxidation known to

date. All three compounds were found to have major

contribution from the PRA mechanism, which can be

attributed to their highly conjugated double bond systems.

Likely due to the presence of a tertiary allylic H atom, vitamin

D₃also has signiﬁcant contribution from HAT.

We then examined the propagation rate constants of

diﬀerent PUFAs, including CLA 18:2, CLA 18:3, and NLA

18:3. As observed in 1,3-CHD and 1,4-CHD, conjugated

PUFAs, including CLA 18:2 and CLA 18:3, have signiﬁcant

contribution from the PRA mechanism while nonconjugated

PUFAs such as NLA 18:3 are k_Hdominant. To our surprise,

conjugated PUFAs appeared to be much more reactive toward

lipid peroxidation compared to their nonconjugated analogues.

For example, the total propagation rate constants for linoleic

acid and CLA 18:2 are 62 and 118 M⁻¹s⁻¹, respectively.

Similarly, CLA 18:3 was found to be highly reactive with a k_p

value of 1235 M⁻¹s⁻¹compared to the k_pvalue of 144 M⁻¹s⁻¹

for NLA 18:3.

DISCUSSION

■

Factors Controlling the Contribution of the HAT and

PRA Reaction to the Overall Propagation Step. In this

study, we report the development of a new peroxyl radical

clock approach that, for the ﬁrst time, simultaneously measures

the rate constants of the HAT and PRA reactions for a variety

of organic and lipid substrates. The structure−reactivity

relationship studies through the application of this new

approach have allowed us to derive ﬁve important factors

that inﬂuence how each mechanism would contribute to the

rate-limiting step of the free radical oxidation mechanism.

(1) Radical stabilization eﬀects are more important than

polar eﬀects in inﬂuencing the reactivities of the PRA reaction.

Through the application of the new peroxyl radical approach to

measure the k_addvalues of para- and meta-substituted styrene

derivatives, we found that only substituents that allow the

delocalization of electron spin via either resonance or

hyperconjugation would signiﬁcantly enhance the reactivities

of PRA reactions. This observation is consistent with

previously reported results that radical stabilization eﬀects

are more important than polar eﬀects in inﬂuencing the

stability of ring-substituted benzylic radicals.^38,44,45In addition,

we found that the experimental k_addvalues and the computed

activation energies of styrene and its para-substituted

derivatives agreed better with the computed spin density

than the atomic polar tensor charge distribution at the benzylic

carbon as seen in Table 1 and Figure 9a and 9b.

Another example for the trend observed here is coenzyme

Q10. Speciﬁcally, the methoxy moiety, which is an electron-

donating substituent, can stabilize the radicals formed after the

I

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

Figure 11. Proposed resonance stabilization of radicals formed after the PRA reaction in coenzyme Q10 by (a) a methoxy group, an electron-

donating substituent, and (b) a carbonyl group, an electron-withdrawing substituent.

a

b

c

Figure 12. Structures and rate constants of diﬀerent nonconjugated and conjugated PUFAs. This work. From ref 22. From ref 24.

PRA reaction at C2 or C3 via resonance (Figure 11a).

Computational modeling of the transition state of peroxyl

radical addition to C2 indicates that the methoxy group, C2,

and C3 appear to be in the same plane in the transition state

(Figure S15). On the other hand, the radicals formed after the

addition of the peroxyl radical into any vinylic carbons can be

delocalized into the carbonyl moiety, which is an electron-

withdrawing substituent (Figure 11b). Thus, the large k_add

value of coenzyme Q10 is further evidence that for the PRA

mechanism, radical stabilization eﬀects are more important

than polar eﬀects.

The small changes in both the measured rate constants and

computed activation energies of the meta-substituted styrenes

supported our observation that polar eﬀects do not contribute

signiﬁcantly to the PRA mechanism. In the transition state of

the addition of peroxyl radical to styrene, there is a

development of both radical characteristic and partial positive

charge at the benzylic carbon. Therefore, we employed the

extended Hammett equation to derive a linear correlation

between the computed activation energies and substituents

eﬀects via the dual radical-cationic Hammett parameters

(Figure 9c). The negative slope of −1.04 conﬁrmed the

predicted reaction mechanism.

(2) The PRA mechanism only contributes signiﬁcantly to

the overall propagation rate constant in structures with

conjugated double bond systems, and the more conjugated

the system is, the larger the increase in the total rate constant.

As an example, due to its conjugated double bond system, 1,3-

CHD has signiﬁcant contribution from the PRA mechanism in

addition to the HAT reaction and, thus, is much more reactive

toward lipid peroxidation compared to its isomer 1,4-CHD,

which is k_Hdominant. The same trend was also observed in

other pairs of isomers investigated in this study, such as NLA

and CLA.

J

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

One of the most surprising and important ﬁndings in this

study is how much more reactive the conjugated double bond

structures are compared to the nonconjugated ones. For

example, the total propagation rate constant measured for CLA

18:3 is 10 times that of NLA 18:3, which is due to the

combination of both PRA and HAT reactions. In addition,

compounds that contain three conjugated double bonds, such

as vitamin D₃and CLA 18:3, were found to have k_pvalues of

over 1000 M⁻¹s⁻¹with signiﬁcant contribution from both

mechanisms. Signiﬁcantly, retinal has a conjugated system with

ﬁve double bonds and thus has a propagation rate constant of

almost 5700 M⁻¹s⁻¹, which is the largest rate constant that has

ever been measured for a lipid molecule.

To fully understand how reactive the conjugated double

bond structures are, we compared the increase in the

propagation rate constants of diﬀerent nonconjugated and

conjugated PUFAs (Figure 12). Nonconjugated PUFAs such

as arachidonic acid, eicosapentaenoic acid, and docosahex-

aenoic acid are commonly considered highly reactive fatty

acids. Their reactivities toward the HAT reaction are attributed

to the number of bis-allylic positions due to the weaker C−H

bonds at these positions.^41,46,47Thus, the ratio of the rate

constants for linoleic acid, NLA 18:3, arachidonic acid,

eicosapentaenoic acid, and docosahexaenoic acid is roughly

1:2:3:4:5, which is proportional to the number of their bis-

allylic positions. However, this increase in rate constant is

much larger for the conjugated double bond systems. For

example, CLA 18:3 has only one more double bond than CLA

18:2, but the diﬀerence in their propagation rate constants is

10 times. Based on the result of retinal, we estimated that a

structure with ﬁve conjugated double bonds would have a k_p

value of roughly 6000 M⁻¹s⁻¹, and a structure with four

conjugated double bonds would have a rate constant in

between 1235 and 6000 M⁻¹s⁻¹.

(3) Monoallylic positions that are adjacent to conjugated

double bond system are reactive toward the HAT reaction.

Monoallylic H atoms in acyclic structures are commonly

believed to be less reactive toward abstraction compared to

those at bis-allylic positions. However, once the values of k_add

and k_Hare distinguished for conjugated PUFAs using the new

peroxyl radical clock, we found that monoallylic positions that

are adjacent to a highly conjugated double bond system are

also highly reactive toward HAT reactions. For example, when

there is an increase in double bond from CLA 18:2 to 18:3,

there is a 6-fold increase in k_Hvalues, though to a lesser extent

than the increase in k_addvalues (14-fold). Our results are, in

fact, consistent with the ﬁndings that C−H BDEs decrease

with extended conjugation of lipids.⁴¹Speciﬁcally, the lowest

theoretical C−H BDEs calculated for lipids with two and three

conjugated double bonds were 77.4 and 72.9 kcal/mol,

respectively, comparable to the bis-allylic C−H bond BDE at

72.7 kcal/mol.⁴¹

(4) Cyclic compounds are more reactive toward lipid

peroxidation than their acyclic analogues, but their k_add/k_H

ratios are the same. The trend that cyclic compounds are more

reactive toward lipid peroxidation than their acyclic analogues

has previously been discussed.⁴Indeed, we found that 1,3-

CHD has a much larger k_pvalue compared to 1,3-hexadiene

and CLA 18:2. However, after measuring the individual k_add

and k_H, we found that all three compounds have almost the

same ratio k_add/k_Hof one. This indicates that both k_Hand k_add

values are larger in the cyclic compounds than their acyclic

analogues.

(5) Steric hindrance and the location of the substituents can

potentially aﬀect the reactivity of the PRA mechanism. Despite

the opposite polar transition state, factors that aﬀect the

reactivities of carbon-radical addition to alkenes can also be

applied to PRA to alkenes. The ﬁrst factor is steric hindrance

which can potentially dictate the regioselectivity of both

reactions. Speciﬁcally, it was found that carbon radicals, and

potentially peroxyl radicals, attack alkenes preferentially at the

carbon atoms which are the least sterically hindered.³⁵The

second factor is the location of the substituents. In the addition

of carbon radical to alkenes, it was found that a substituent that

is bound to the α-carbon, to which the radical is added, exerts

markedly diﬀerent eﬀects compared to a substituent attached

to the neighboring carbon (the β-carbon);³⁵i.e., the β-

substituent only exerts a polar eﬀect on the rate constant while

the α-substituent mostly exerts both polar and steric eﬀects,

but the polar eﬀect is less than that of the β-substituent.

Therefore, we propose that the same principle can be applied

to the reaction between peroxyl radical and alkenes. In fact,

even though 1,3-hexadiene, 2,4-hexadiene, and 3-methyl-1,3-

pentadiene all have an additional methyl group compared to

1,3- pentadiene, only 2,4-hexadiene and 3-methyl-1,3-

pentadiene have a substantial increase in k_adddue to the

polar eﬀect of the additional methyl group on the conjugated

double bonds. Replacing the methyl with an ethyl group in 1,3-

hexadiene and the addition of a methyl on the other end of the

alkene in 2,4-hexadiene resulted in substantial k_H, but addition

of the methyl at the cross-conjugation position of the alkene in

3-methyl-1,3-pentadiene did not result in an observable k_H.

Furthermore, addition of two more terminal methyl groups in

2,5-dimethyl-2,4-hexadiene resulted in large increases in both

k_addand k_H. On the other hand, the smaller k_addvalue of 2,4-

dimethyl-1,3- pentadiene relative to 3-methyl-1,3-pentadiene is

likely due to the steric hindrance at one of the terminal vinyl

carbons. However, for this series of dienes overall, the steric

eﬀect appears to be less important than the radical-stabilizing

eﬀect of the additional substituents. Further investigation is

required to examine whether a bulkier substituent on the α-

carbon could exert a larger steric eﬀect for the PRA reactions.

Limitations of the New Peroxyl Radical Clock

Approach. We found that there is a threshold to which the

new peroxyl radical clock approach can distinguish between k_H

and k_addvalues. Since a PRA-derived epoxide product, 7-DHC

5α,6α-epoxide, was previously detected among the oxidation

products of 7-DHC,^12,29,30we predicted that there should be

signiﬁcant contribution from the PRA mechanism to the

peroxidation of 7-DHC. However, our results showed that 7-

DHC is k_Hdominant. This ﬁnding is surprising as we

mentioned that structures with conjugated double bonds are

expected to have a large k_addvalue. We attempted to resolve

this conﬂict in results by investigating α-terpinene, a

compound that shares a cyclic diene and a tertiary monoallylic

position with the 7-DHC structure (Figure 4e and Table 2). In

fact, the abstraction of an H atom from the tertiary monoallylic

position of α-terpinene would yield a highly stable carbon

radical. This explains the large increase in both the total

propagation and HAT rate constants compared to 1,3-CHD,

which only has the cyclic diene in its structure. Signiﬁcantly,

the k_add/k_Hratio also decreases from 0.95 in 1,3-CHD to 0.28

in α-terpinene. Compared to α-terpinene, 7-DHC has two

tertiary monoallylic positions, a much more rigid structure, and

a well-aligned axial C−H bond with the π-orbitals, which

requires minimum reorientation of the molecule to reach the

K

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

transition of HAT.^24,30Thus, it is reasonable to see that 7-

DHC displays a much larger k_Hthan α-terpinene. On the other

hand, a cholestadienol with a similar substitution pattern as α-

terpinene has a k_pof 1370 M⁻¹s⁻¹, similar to that of α-

terpinene.⁴⁸Although there is contribution from the PRA

mechanism to the total propagation rate constant of 7-DHC,

the k_addvalue may be too small to be detected by the new

radical clock, which relies on accurate determination of the

components by RSL3 and erastin, respectively, leads to the

accumulation of lipid oxidation products via the free radical

chain reactions to lethal levels, resulting in cell death, i.e.,

ferroptosis.¹⁶Since lipid peroxidation plays a central role in the

execution of ferroptosis, nonconjugated PUFAs, such as

arachidonic acid, due to their high reactivities toward lipid

peroxidation, have been found to potentiate ferroptosis.^61,63,64

Based on the surprisingly large k_pvalue measured in this study

for conjugated PUFAs, we suggest that the cytotoxicity of

conjugated PUFAs to cancer cells can be attributed to its high

reactivity toward lipid peroxidation. The ability of conjugated

PUFAs to enhance ferroptosis could also be related to their

oxidation product distribution, which should be composed of

large amounts of addition products, i.e., allylic epoxides. These

epoxides are highly electrophilic and can form adducts with

macromolecules, such as proteins and DNA, leading to

damages.^65,66Besides conjugated PUFAs, the peroxidation

reactivities and mechanism of coenzyme Q10 and lipophilic

vitamins, such as retinal and vitamin D₃, were determined for

the ﬁrst time in this study. Coenzyme Q10, retinal, and vitamin

D₃have been found to play important roles in a variety of

biological processes.⁶⁷⁻⁶⁹However, only the reactivities and

peroxidation mechanism of vitamin A precursor, β-carotene,

have been studied previously.⁷⁰⁻⁷⁵It would be intriguing to

explore how the high reactivities of these lipid species would

translate into their biological activities, especially in ferroptosis.

The exceptionally large rate constant of vitamin A, retinal in

this study, is of particular interest because it has been reported

to act as both an antioxidant and a pro-oxidant under low and

high oxygen tension, respectively.⁷⁰⁻⁷³Considering the fact

that oxygen tension varies signiﬁcantly in diﬀerent tissues/

organs, vitamin A might play diﬀerent roles in inducing or

inhibiting ferroptosis in diﬀerent types of cancer cells.

k

_add/k_Hratio. Similarly, even though 12% of the total oxidation

products from cholesterol or cholesteryl acetate were addition

products,²⁹our new approach was not sensitive enough to

measure the k_addof this substrate. Therefore, we concluded

that there is a threshold of the k_add/k_Hratio at which the new

radical clock approach can resolve the individual k_addand k_H

values. In the case where the contribution from one

mechanism is signiﬁcantly less than the other, the new

approach is not sensitive enough to resolve them. The lowest

k

_add/k_Hratio that we were able to measure in this study was

0.28 of α-terpinene. Improvement on the new radical clock

approach would be needed in the future to overcome this

limitation.

Biological Implications of the Lipid Peroxidation

Mechanism. The diﬀerentiation between HAT and PRA

reactions in the propagation step will allow us to better

understand not only the lipid peroxidation kinetics and

mechanism but also the distribution of the oxidation products

of a lipid, which translates into its biological and pathological

eﬀects. For example, a lipid that primarily undergoes the PRA

mechanism would yield more addition products such as

epoxides and, thus, would exert very diﬀerent biological eﬀects

compared to a lipid that predominantly undergoes the HAT

mechanism, which generates mostly lipid hydroperoxides.

Therefore, it would be interesting to compare the biological

eﬀects of nonconjugated PUFAs, which predominantly under-

go the HAT reaction, and conjugated PUFAs, which can also

undergo PRA in addition to the HAT reaction. Even though

the peroxidation reactivities and mechanism of nonconjugated

PUFAs, such as arachidonic acid, have been studied

extensively,^2,49to our knowledge, this is the ﬁrst study that

explored the reactivities and mechanism toward lipid

peroxidation of conjugated PUFAs, including CLA 18:2 and

CLA 18:3. Conjugated linoleic and linolenic acids do not occur

naturally in humans and can only be found in meat and dairy

products from ruminants and plant-based food products such

as pomegranate seed oil and bitter gourd seed.⁵⁰⁻⁵³However,

these conjugated fatty acids have attracted many research

eﬀorts due to their health-promoting properties, such as

antidiabetic,⁵⁴anti-inﬂammatory,⁵⁵antiatherogenic,⁵⁶and

anticarcinogen.^57,58

Interestingly, parinaric acid, an 18:4 conjugated fatty acid, is

>25 times more cytotoxic to cancer cells than the

corresponding nonconjugated fatty acids, and the antioxidant,

butylated hydroxytoluene, abolishes its cytotoxicity, indicating

a lipid peroxidation-mediated cell death mechanism.⁵⁹Another

CLA 18:3, α-eleostearic acid, was recently found to

signiﬁcantly sensitize triple-negative breast cancer cells to

ferroptosis,⁶⁰a type of cell death that is tightly associated with

lipid peroxidation.^13,14,61Due to the substantial generation of

ROS and constantly being under high levels of oxidative

stress,⁶²certain types of cancer cell lines rely heavily on the

functions of the components in the cellular antioxidant

network, including glutathione peroxidase 4 and system x_c⁻,

for their survival.^13,14Therefore, inhibition of these two

EXPERIMENTAL SECTION

■

General Methods and Materials. The radical initiator,

MeOAMVN, was purchased from Wako Chemicals, dried under

vacuum, and stored at −80 °C. MeLn (Nu-Chek-Prep, Inc.) was

puriﬁed through silica gel (10% ethyl acetate:90% hexane) prior to

use and stored at −80 °C. Sty, 4-MeSty, 4-MeOSty, 4-ClSty, 4-FSty,

3-MeSty, 3-MeOSty, 3-ClSty, α-MeSty, (trans-)β-MeSty, 1,3-

pentadiene, 3-methyl-1,3-pentadiene, 2,4-dimethyl-1,3-pentadiene,

1,3-hexadiene, 2,4-hexadiene, 2,5-dimethyl-2,4-hexadiene, 1,2-dihy-

dronaphthalene, indene, 2-methylindene, and α-terpinene were

purchased from Sigma-Aldrich Co. and were puriﬁed through a

short Al₂O₃column to remove stabilizers such as 4-tert-butylcatechol

prior to use. 4-NO₂Sty was purchased from Thermo Fisher Scientiﬁc

Inc., stored at −20 °C, and puriﬁed through a column to remove

stabilizers prior to use. 7-DHC was purchased from Sigma-Aldrich

Co., stored at −20 °C, and puriﬁed through silica gel column prior to

use. Cholesteryl acetate was purchased from Thermo Fisher Scientiﬁc

Inc. and used without further puriﬁcation. Retinal, coenzyme Q10,

and Vitamin D₃were purchased from Chem-Impex Int’l Inc. and used

without further puriﬁcation. CLA 18:2 and NLA 18:3 were purchased

from Nu-Chek-Prep, Inc. and used without further puriﬁcation.

9(E),11(E),13(E)-CLA 18:3 (≥97%) was purchased from Cayman

Chemical and used without further puriﬁcation. The internal

standard, 9-anthracenemethanol (97%), was purchased from Sigma-

Aldrich Co. LiAlH₄(1 M in THF) was purchased from Sigma-Aldrich

Co. Benzene (HPLC grade) was passed through a column of neutral

alumina prior to use. HPLC grade hexanes and isopropyl alcohol were

purchased from Thermo Fisher Scientiﬁc Inc.

General Procedure for the New Peroxyl Radical Clock

Approach Using Methyl Linoleate. The initial experimental setup

for the new approach is similar to the conventional peroxyl radical

clock method.²³Brieﬂy, for each reaction, stock solution of the

L

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

substrate at various amounts was added ﬁrst to each reaction vial.

Then, the same amount of stock solution of MeLn in benzene, an

adjusted amount of benzene, and the same amount of stock solution

of MeOAMVN in benzene were added to bring the ﬁnal reaction

volume in each vial to 300 μL. The ﬁnal concentrations of MeLn and

MeOAMVN are 0.302 M and 1.5 mM, respectively. Each reaction was

carried out at 37 °C for an hour and then quenched by adding

butylated hydroxytoluene (50 μL, 0.2 M in benzene) and 9-

anthracenemethanol (6 μL, 3 mM in chloroform). The resulting

mixture was vortexed for 5 s and left at room temperature for 30 min

before splitting into two portions of the same volume. The ﬁrst

portion was diluted with 1 mL of hexane for the analysis of HPODEs

by HPLC-UV (Phenomenex Luna 4.6 × 150 mm Si column; 3 μm

particle size; 1.0 mL/min; eluting solvent, Hex:Iso = 99.5:0.5). The

other half of the reaction mixture was reduced by the addition of

LiAlH₄(90 μL, 1 M in THF). The mixture was then vortexed for 5 s,

left at room temperature for 15 min, and quenched by the addition of

concentrated HCl (20 μL). The ﬁnal reaction mixture was diluted

with 1 mL of hexane, followed by ﬁltration with Teﬂon syringe ﬁlter,

for the analysis of HODOLs by HPLC-UV with eluting solvent of

Hex:Iso = 95.5:4.5 using the same column. The measurement of rate

constants for each substrate is performed in triplicate. The peak areas

for HPODEs and HODOLs were integrated at 234 nm. In addition,

the peak areas for the internal standard were integrated at 254 nm.

The ratio [t,c-HODOLs]/[t,t-HODOLs] is calculated as below:

mL/min; eluting solvent, Hex:Iso = 95.5:4.5) to obtain individual

HODOLs. Ag⁺-coordination electrospray ionization analysis of

individual HODOLs on a high-resolution QTOF mass spectrometer

(Waters Synapt XS) revealed expected doublets due to the isotopes of

Ag (¹⁰⁷Ag and ¹⁰⁹Ag) (Figure S14).

1

13-trans, trans-HODOL. H NMR (CDCl₃, 300 MHz): 0.88 (t,

3H, J = 6.8 Hz), 1.30 and 1.56 (br s, 20H), 2.07 (q, 2H, J = 6.9 Hz),

3.64 (t, 2H, J = 6.6 Hz), 4.11 (q, 1H, J = 6.5 Hz), 5.57 (dd, 1H, J =

7.0 and 15.1 Hz), 5.70 (dt, 1H, J = 7.2 and 14.9 Hz), 6.02 (dd, 1H, J

= 10.4 and 15.0 Hz), 6.17 (dd, 1H, J = 10.4 and 15.2 Hz). ¹³C NMR

(CDCl₃, 500 MHz): 14.06, 22.62, 25.15, 25.72, 29.09, 29.19, 29.36,

29.41, 31.81, 32.63, 32.81, 37.33, 63.10, 72.93, 129.50, 130.96, 133.69,

135.51. HRMS [M + ¹⁰⁷Ag]⁺(C₁₈H₃₄AgO₂): observed, 389.1652;

theoretical, 389.1604. UV−vis: λ_max(in hexanes) = 230.9 nm.

9-trans, trans-HODOL. ¹H NMR (CDCl₃): ¹H NMR (CDCl₃, 300

MHz): 0.89 (t, 3H, J = 6.8 Hz), 1.31 and 1.56 (br s, 20H), 2.07 (q,

2H, J = 6.9 Hz), 3.64 (t, 2H, J = 6.6 Hz), 4.10 (q, 1H, J = 6.5 Hz),

5.57 (dd, 1H, J = 7.0 and 15.0 Hz), 5.70 (dt, 1H, J = 7.2 and 14.9

Hz), 6.02 (dd, 1H, J = 10.4 and 14.9 Hz), 6.17 (dd, 1H, J = 10.4 and

14.9 Hz). ¹³C NMR (CDCl₃, 500 MHz): 14.06, 22.54, 25.43, 25.72,

28.93, 29.36, 29.50, 29.52, 31.43, 32.63, 32.81, 37.33, 63.10, 72.91,

129.42, 131.01, 133.61, 135.66. HRMS [M + ¹⁰⁷Ag]⁺(C₁₈H₃₄AgO₂):

observed, 389.1652; theoretical, 389.1604. UV−vis: λ_max(in hexanes)

= 230.9 nm.

13-trans, cis-HODOL. HRMS [M + ¹⁰⁷Ag]⁺(C₁₈H₃₄AgO₂):

observed, 389.1639; theoretical, 389.1604. UV−vis: λ_max(in hexanes)

= 234.4 nm.

(total areas under all t, c‐HODOLs)

(total areas under all t, t‐HODOLs)

[t, c‐HODOLs]

[t, t‐HODOLs]

=

9-trans, cis-HODOL. HRMS [M + ¹⁰⁷Ag]⁺(C₁₈H₃₄AgO₂):

observed, 389.1644; theoretical, 389.1604. UV−vis: λ_max(in hexanes)

= 234.4 nm.

The ratio [HODOL_total]/[HPODE_total] is calculated as below:

[HODOL_total

[HPODE_total

Computational Method. All geometry optimization was

performed using the Gaussian 16 software⁷⁶with the hybrid density

functional B3LYP and the 6-31++G(d,p) basis. Transition state

optimization and calculations, including activation energies, spin

densities, and atomic polar tensor charge distribution, were also

carried out using the B3LYP/6-31++G(d,p) with QST2 method. The

activation energies were calculated by deducting the sum of total

energies (electronic + thermal free energies) of styrene derivatives and

methyl peroxyl radicals from the total energies (electronic + thermal

free energies) of the transition state.

]

(total areas under all HODOLs)/(total area of internal standard)

(total areas under all HPODEs)/(total area of internal standard)

=

General Procedure for Free Radical Oxidation of 7-DHC

with the New Peroxyl Radical Clock Approach. The oxidation,

reduction, and HPLC-UV analysis for 7-DHC were carried out

similarly as described above. However, the stock solution of 7-DHC

was prepared in a solvent mixture of an equal amount of

chlorobenzene and benzene (chlorobenzene:benzene = 1:1). The

reaction was also carried out in chlorobenzene:benzene to increase

the solubility of 7-DHC.

General Procedure for Free Radical Oxidation of Retinal

with the New Peroxyl Radical Clock Approach. The oxidation,

reduction, and HPLC-UV analysis for retinal were carried out

similarly as described above, but the experiment was carried out in

only 30 min instead of 1 h.

General Procedure for Free Radical Oxidation of NLA 18:2,

CLA 18:2, and CLA 18:3 with the New Peroxyl Radical Clock

Approach. The oxidation, reduction, and HPLC-UV analysis for

NLA 18:2, CLA 18:2, and CLA 18:3 were carried out similarly as

described above. However, the −OOH moiety in these three fatty

acids gives a broad peak in the HPLC chromatograms that overlaps

with the HPODEs product peaks. Therefore, to avoid this

overlapping, the nonreduced reaction mixture of these three

compounds was analyzed using an eluting solvent of Hex:Iso =

99.864:0.136.

ASSOCIATED CONTENT

* Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.joc.0c01920.

■

sı

Experimental data (XLSX)

Supplementary data of k_addand k_Hcalculations of all

substrates; HPLC chromatograms of representative

substrates; characterization of HODOL products,

including HPLC/UV, mass spectrometry, and NMR

data; computational modeling of the addition of methyl

peroxyl radical to coenzyme Q₁₀; supplementary data of

the generation of Hammett plots; activation energies

and Cartesian coordinates for styrene and its derivatives

(PDF)

Preparation of HODOLs. A large-scale autoxidation was carried

out using 5 g of methyl linoleate and 5 mol % of MeOAMVN at 37

°C. The product mixture was then quenched by the addition of BHT

and then reduced with LiAlH₄at 0 °C. After an excess amount of

LiAlH₄was quenched with concentrated HCl, the solution was

neutralized by the addition of saturated NaHCO₃solution until there

were no bubbles. The resulting solution was then extracted with

methylene chloride and ethyl acetate and concentrated. The product

mixture was then ﬁrst puriﬁed by ﬂash column chromatography to

enrich HODOL-containing fractions and remove the majority of

reduced methyl linoleate. Part of the HODOL mixture was then

subject to puriﬁcation by HPLC (250 × 10 mm Si column; 5 μm; 6

AUTHOR INFORMATION

Corresponding Author

■

Libin Xu − Department of Medicinal Chemistry, University of

Washington, Seattle, Washington 98195, United States;

orcid.org/0000-0003-1021-5200; Phone: (206) 543-

1080; Email: libinxu@uw.edu; Fax: (206) 685-3252

Authors

Quynh Do − Department of Medicinal Chemistry, University

of Washington, Seattle, Washington 98195, United States

M

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

iron-dependent form of nonapoptotic cell death. Cell 2012, 149 (5),

1060−72.

(14) Yang, W. S.; SriRamaratnam, R.; Welsch, M. E.; Shimada, K.;

Skouta, R.; Viswanathan, V. S.; Cheah, J. H.; Clemons, P. A.; Shamji,

A. F.; Clish, C. B.; Brown, L. M.; Girotti, A. W.; Cornish, V. W.;

Schreiber, S. L.; Stockwell, B. R. Regulation of ferroptotic cancer cell

death by GPX4. Cell 2014, 156 (1−2), 317−331.

(15) Zilka, O.; Shah, R.; Li, B.; Friedmann Angeli, J. P.; Griesser, M.;

Conrad, M.; Pratt, D. A. On the Mechanism of Cytoprotection by

Ferrostatin-1 and Liproxstatin-1 and the Role of Lipid Peroxidation in

Ferroptotic Cell Death. ACS Cent. Sci. 2017, 3 (3), 232−243.

(16) Shah, R.; Shchepinov, M. S.; Pratt, D. A. Resolving the Role of

Lipoxygenases in the Initiation and Execution of Ferroptosis. ACS

Cent. Sci. 2018, 4 (3), 387−396.

David D. Lee − Department of Medicinal Chemistry,

University of Washington, Seattle, Washington 98195,

United States

Andrew N. Dinh − Department of Medicinal Chemistry,

University of Washington, Seattle, Washington 98195,

United States; orcid.org/0000-0002-9708-4583

Ryan P. Seguin − Department of Medicinal Chemistry,

University of Washington, Seattle, Washington 98195,

United States

Rutan Zhang − Department of Medicinal Chemistry,

University of Washington, Seattle, Washington 98195,

United States

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.joc.0c01920

(17) Conrad, M.; Pratt, D. A. The chemical basis of ferroptosis. Nat.

Chem. Biol. 2019, 15 (12), 1137−1147.

(18) Porter, N. A. A Perspective on Free Radical Autoxidation: The

Physical Organic Chemistry of Polyunsaturated Fatty Acid and Sterol

Peroxidation. J. Org. Chem. 2013, 78 (8), 3511−3524.

Notes

The authors declare no competing ﬁnancial interest.

(19) Zielinski, Z. A. M.; Pratt, D. A. Lipid Peroxidation: Kinetics,

Mechanisms, and Products. J. Org. Chem. 2017, 82 (6), 2817−2825.

(20) Howard, J.; Ingold, K. Absolute Rate Constants for Hydro-

carbon Autoxidation: IV. Tetralin, Cyclohexene, Diphenylmethane,

Ethylbenzene, And Allylbenzene. Can. J. Chem. 1966, 44, 1119−1130.

(21) Howard, J. A.; Ingold, K. U.; Symonds, M. Absolute rate

constants for hydrocarbon oxidation. VIII. The reactions of

cumylperoxy radicals. Can. J. Chem. 1968, 46 (6), 1017−1022.

(22) Howard, J.; Ingold, K. Absolute rate constants for hydrocarbon

autoxidation. VI. Alkyl aromatic and olefinic hydrocarbons. Can. J.

Chem. 1967, 45, 793−802.

ACKNOWLEDGMENTS

■

This work is supported by a grant from the National Science

Foundation (CHE-1664851). The authors thank Dylan H.

Ross for helping with setting up the transition state calculations

using Gaussian.

REFERENCES

■

(1) Ray, P. D.; Huang, B. W.; Tsuji, Y. Reactive oxygen species

(ROS) homeostasis and redox regulation in cellular signaling. Cell.

Signalling 2012, 24 (5), 981−90.

(2) Yin, H.; Xu, L.; Porter, N. A. Free Radical Lipid Peroxidation:

Mechanisms and Analysis. Chem. Rev. 2011, 111 (10), 5944−5972.

(3) Halliwell, B.; Chirico, S. Lipid peroxidation: Its mechanism,

measurement, and significance. Am. J. Clin. Nutr. 1993, 57 (5), 715S−

725S.

(4) Xu, L.; Porter, N. A. Free radical oxidation of cholesterol and its

precursors: Implications in cholesterol biosynthesis disorders. Free

Radical Res. 2015, 49 (7), 835−849.

(5) Brown, A. J.; Jessup, W. Oxysterols and atherosclerosis.

Atherosclerosis 1999, 142 (1), 1−28.

(6) Davi, G.; Falco, A.; Patrono, C. Lipid peroxidation in diabetes

mellitus. Antioxid. Redox Signaling 2005, 7 (1−2), 256−68.

(7) Ames, B. N. DNA damage from micronutrient deficiencies is

likely to be a major cause of cancer. Mutat. Res., Fundam. Mol. Mech.

Mutagen. 2001, 475 (1−2), 7−20.

(8) Barrera, G. Oxidative stress and lipid peroxidation products in

cancer progression and therapy. ISRN Oncol 2012, 2012, 137289.

(9) Simonian, N. A.; Coyle, J. T. Oxidative stress in neuro-

degenerative diseases. Annu. Rev. Pharmacol. Toxicol. 1996, 36, 83−

106.

(10) Sayre, L. M.; Zelasko, D. A.; Harris, P. L.; Perry, G.; Salomon,

R. G.; Smith, M. A. 4-Hydroxynonenal-derived advanced lipid

peroxidation end products are increased in Alzheimer ’s disease. J.

Neurochem. 1997, 68 (5), 2092−7.

(11) Porter, F. D.; Scherrer, D. E.; Lanier, M. H.; Langmade, S. J.;

Molugu, V.; Gale, S. E.; Olzeski, D.; Sidhu, R.; Dietzen, D. J.; Fu, R.;

Wassif, C. A.; Yanjanin, N. M.; Marso, S. P.; House, J.; Vite, C.;

Schaffer, J. E.; Ory, D. S. Cholesterol oxidation products are sensitive

and specific blood-based biomarkers for Niemann-Pick C1 disease.

Sci. Transl. Med. 2010, 2 (56), 56ra81.

(23) Roschek, B.; Tallman, K. A.; Rector, C. L.; Gillmore, J. G.;

Pratt, D. A.; Punta, C.; Porter, N. A. Peroxyl Radical Clocks. J. Org.

Chem. 2006, 71 (9), 3527−3532.

(24) Xu, L.; Davis, T. A.; Porter, N. A. Rate Constants for

Peroxidation of Polyunsaturated Fatty Acids and Sterols in Solution

and in Liposomes. J. Am. Chem. Soc. 2009, 131 (36), 13037−13044.

(25) Lamberson, C. R.; Muchalski, H.; McDuffee, K. B.; Tallman, K.

A.; Xu, L.; Porter, N. A. Propagation rate constants for the

peroxidation of sterols on the biosynthetic pathway to cholesterol.

Chem. Phys. Lipids 2017, 207, 51−58.

(26) Pratt, D. A.; Tallman, K. A.; Porter, N. A. Free Radical

Oxidation of Polyunsaturated Lipids: New Mechanistic Insights and

the Development of Peroxyl Radical Clocks. Acc. Chem. Res. 2011, 44

(6), 458−467.

(27) Kresge, N.; Simoni, R. D.; Hill, R. L. The Biosynthetic Pathway

for Cholesterol: Konrad Bloch. J. Biol. Chem. 2005, 280 (10), e7.

(28) Holick, M. F.; Frommer, J. E.; McNeill, S. C.; Richtand, N. M.;

Henley, J. W.; Potts, J. T., Jr. Photometabolism of 7-dehydrocholes-

terol to previtamin D3 in skin. Biochem. Biophys. Res. Commun. 1977,

76 (1), 107−14.

(29) Zielinski, Z. A. M.; Pratt, D. A. H-Atom Abstraction vs

Addition: Accounting for the Diverse Product Distribution in the

Autoxidation of Cholesterol and Its Esters. J. Am. Chem. Soc. 2019,

141 (7), 3037−3051.

(30) Xu, L.; Korade, Z.; Porter, N. A. Oxysterols from Free Radical

Chain Oxidation of 7-Dehydrocholesterol: Product and Mechanistic

Studies. J. Am. Chem. Soc. 2010, 132 (7), 2222−2232.

(31) Liu, W.; Porter, N. A.; Schneider, C.; Brash, A. R.; Yin, H.

Formation of 4-hydroxynonenal from cardiolipin oxidation: Intra-

molecular peroxyl radical addition and decomposition. Free Radical

Biol. Med. 2011, 50 (1), 166−178.

(32) Baschieri, A.; Pizzol, R.; Guo, Y.; Amorati, R.; Valgimigli, L.

Calibration of Squalene, p-Cymene, and Sunflower Oil as Standard

Oxidizable Substrates for Quantitative Antioxidant Testing. J. Agric.

Food Chem. 2019, 67 (24), 6902−6910.

(33) Howard, J. A.; Ingold, K. U. Absolute Rate Constants for

Hydrocarbon AutoxidatioN: I. Styrene. Can. J. Chem. 1965, 43 (10),

2729−2736.

(12) Xu, L.; Korade, Z.; Rosado, D. A.; Liu, W.; Lamberson, C. R.;

Porter, N. A. An oxysterol biomarker for 7-dehydrocholesterol

oxidation in cell/mouse models for Smith-Lemli-Opitz syndrome. J.

Lipid Res. 2011, 52 (6), 1222−1233.

(13) Dixon, S. J.; Lemberg, K. M.; Lamprecht, M. R.; Skouta, R.;

Zaitsev, E. M.; Gleason, C. E.; Patel, D. N.; Bauer, A. J.; Cantley, A.

M.; Yang, W. S.; Morrison, B., 3rd; Stockwell, B. R. Ferroptosis: an

N

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

(34) Stark, M. S. Addition of Peroxyl Radicals to Alkenes and the

Reaction of Oxygen with Alkyl Radicals. J. Am. Chem. Soc. 2000, 122

(17), 4162−4170.

conjugated linoleic acid may be mediated via anti-inflammatory effects

in white adipose tissue. Diabetes 2007, 56 (3), 574−82.

(55) Dipasquale, D.; Basirico, L.; Morera, P.; Primi, R.; Troscher, A.;

Bernabucci, U. Anti-inflammatory effects of conjugated linoleic acid

isomers and essential fatty acids in bovine mammary epithelial cells.

Animal 2018, 12 (10), 2108−2114.

(56) Nicolosi, R. J.; Rogers, E. J.; Kritchevsky, D.; Scimeca, J. A.;

Huth, P. J. Dietary conjugated linoleic acid reduces plasma

lipoproteins and early aortic atherosclerosis in hypercholesterolemic

hamsters. Artery 1997, 22 (5), 266−77.

(57) Scimeca, J. A.; Thompson, H. J.; Ip, C. Eﬀect of Conjugated

Linoleic Acid on Carcinogenesis. In Diet and Breast Cancer;

Weisburger, E. K., Ed.; Springer US: Boston, MA, 1994; pp 59−65.

(58) Igarashi, M.; Miyazawa, T. Newly recognized cytotoxic effect of

conjugated trienoic fatty acids on cultured human tumor cells. Cancer

Lett. 2000, 148 (2), 173−9.

(59) Cornelius, A. S.; Yerram, N. R.; Kratz, D. A.; Spector, A. A.

Cytotoxic effect of cis-parinaric acid in cultured malignant cells.

Cancer Res. 1991, 51 (22), 6025−6030.

(60) Beatty, A.; Singh, T.; Tyurina, Y. Y.; Nicolas, E.; Maslar, K.;

Zhou, Y.; Cai, K. Q.; Tan, Y.; Doll, S.; Conrad, M.; Bayır, H.; Kagan,

V. E.; Rennefahrt, U.; Peterson, J. R. Conjugated linolenic fatty acids

trigger ferroptosis in triple-negative breast cancer. bioRxiv 2019,

556084.

(61) Yang, W. S.; Kim, K. J.; Gaschler, M. M.; Patel, M.; Shchepinov,

M. S.; Stockwell, B. R. Peroxidation of polyunsaturated fatty acids by

lipoxygenases drives ferroptosis. Proc. Natl. Acad. Sci. U. S. A. 2016,

113 (34), E4966−75.

(62) Szatrowski, T. P.; Nathan, C. F. Production of Large Amounts

of Hydrogen Peroxide by Human Tumor Cells. Cancer Res. 1991, 51

(3), 794.

(35) Giese, B. Formation of CC Bonds by Addition of Free Radicals

to Alkenes. Angew. Chem., Int. Ed. Engl. 1983, 22 (10), 753−764.

(36) McDaniel, D. H.; Brown, H. C. An Extended Table of

Hammett Substitutent Constants Based on the Ionization of

Substituted Benzoic Acids. J. Org. Chem. 1958, 23 (3), 420−427.

(37) Fisher, T. H.; Meierhoefer, A. W. Substituent effects in free-

radical reactions. A study of 4-substituted 3-cyanobenzyl free radicals.

J. Org. Chem. 1978, 43 (2), 224−228.

(38) Dincţ urk, S.; Jackson, R. A.; Townson, M. An improved σ ·

̈

scale. The thermal decomposition of substituted dibenzylmercury

compounds in alkane solutions. J. Chem. Soc., Chem. Commun. 1979,

4, 172−174.

(39) Taylor, R. Electrophilic aromatic substitution; John Wiley:

Chichester, England; New York, 1990; pp xvi, 513.

(40) Porter, N. A.; Lehman, L. S.; Weber, B. A.; Smith, K. J. Unified

mechanism for polyunsaturated fatty acid autoxidation. Competition

of peroxy radical hydrogen atom abstraction, .beta.-scission, and

cyclization. J. Am. Chem. Soc. 1981, 103 (21), 6447−6455.

(41) Pratt, D. A.; Mills, J. H.; Porter, N. A. Theoretical Calculations

of Carbon −Oxygen Bond Dissociation Enthalpies of Peroxyl Radicals

Formed in the Autoxidation of Lipids. J. Am. Chem. Soc. 2003, 125

(19), 5801−5810.

(42) Korcek, S.; Chenier, J.; Howard, J.; Ingold, K. Absolute Rate

Constants for Hydrocarbon Autoxidation. XXI. Activation Energies

for Propagation and the Correlation of Propagation Rate Constants

with Carbon −Hydrogen Bond Strengths. Can. J. Chem. 1972, 50,

2285−2297.

(43) Howard, J. A.; Ingold, K. U. Absolute rate constants for

hydrocarbon autoxidation. V. The hydroperoxy radical in chain

propagation and termination. Can. J. Chem. 1967, 45 (8), 785−792.

(44) Pratt, D. A.; DiLabio, G. A.; Mulder, P.; Ingold, K. U. Bond

Strengths of Toluenes, Anilines, and Phenols: To Hammett or Not.

Acc. Chem. Res. 2004, 37 (5), 334−340.

(45) Bathelt, C. M.; Ridder, L.; Mulholland, A. J.; Harvey, J. N.

Aromatic Hydroxylation by Cytochrome P450: Model Calculations of

Mechanism and Substituent Effects. J. Am. Chem. Soc. 2003, 125 (49),

15004−15005.

(46) Trenwith, A. B. Dissociation of 3-methylpenta-1, 4-diene and

the resonance energy of the pentadienyl radical. J. Chem. Soc., Faraday

Trans. 1 1982, 78 (10), 3131−3136.

(47) Berkowitz, J.; Ellison, G. B.; Gutman, D. Three methods to

measure RH bond energies. J. Phys. Chem. 1994, 98 (11), 2744−2765.

(48) Xu, L.; Porter, N. A. Reactivities and Products of Free Radical

Oxidation of Cholestadienols. J. Am. Chem. Soc. 2014, 136 (14),

5443−5450.

(49) Aliwarga, T.; Raccor, B. S.; Lemaitre, R. N.; Sotoodehnia, N.;

Gharib, S. A.; Xu, L.; Totah, R. A. Enzymatic and free radical

formation of cis- and trans- epoxyeicosatrienoic acids in vitro and in

vivo. Free Radical Biol. Med. 2017, 112, 131−140.

(50) Dhiman, T. R.; Anand, G. R.; Satter, L. D.; Pariza, M. W.

Conjugated linoleic acid content of milk from cows fed different diets.

J. Dairy Sci. 1999, 82 (10), 2146−56.

(51) Destaillats, F.; Trottier, J. P.; Galvez, J. M.; Angers, P. Analysis

of alpha-linolenic acid biohydrogenation intermediates in milk fat

with emphasis on conjugated linolenic acids. J. Dairy Sci. 2005, 88

(9), 3231−9.

(52) Suzuki, R.; Noguchi, R.; Ota, T.; Abe, M.; Miyashita, K.;

Kawada, T. Cytotoxic effect of conjugated trienoic fatty acids on

mouse tumor and human monocytic leukemia cells. Lipids 2001, 36

(5), 477−82.

(63) Doll, S.; Proneth, B.; Tyurina, Y. Y.; Panzilius, E.; Kobayashi,

S.; Ingold, I.; Irmler, M.; Beckers, J.; Aichler, M.; Walch, A.; Prokisch,

̈ ̈

H.; Trumbach, D.; Mao, G.; Qu, F.; Bayir, H.; Fullekrug, J.; Scheel, C.

H.; Wurst, W.; Schick, J. A.; Kagan, V. E.; Angeli, J. P. F.; Conrad, M.

ACSL4 dictates ferroptosis sensitivity by shaping cellular lipid

composition. Nat. Chem. Biol. 2017, 13 (1), 91−98.

(64) Dixon, S. J.; Winter, G. E.; Musavi, L. S.; Lee, E. D.; Snijder, B.;

Rebsamen, M.; Superti-Furga, G.; Stockwell, B. R. Human Haploid

Cell Genetics Reveals Roles for Lipid Metabolism Genes in

Nonapoptotic Cell Death. ACS Chem. Biol. 2015, 10 (7), 1604−1609.

(65) Zhang, J.; Wang, C.; Ji, L.; Liu, W. Modeling of Toxicity-

Relevant Electrophilic Reactivity for Guanine with Epoxides:

Estimating the Hard and Soft Acids and Bases (HSAB) Parameter

as a Predictor. Chem. Res. Toxicol. 2016, 29 (5), 841−850.

́

(66) Koskinen, M.; Plna, K. Specific DNA adducts induced by some

mono-substitued epoxides in vitro and in vivo. Chem.-Biol. Interact.

2000, 129 (3), 209−229.

(67) Zile, M. H. Function of vitamin A in vertebrate embryonic

development. J. Nutr. 2001, 131 (3), 705−8.

(68) Holick, M. F. Vitamin D: importance in the prevention of

cancers, type 1 diabetes, heart disease, and osteoporosis. Am. J. Clin.

Nutr. 2004, 79 (3), 362−71.

(69) Lenaz, G.; Fato, R.; Formiggini, G.; Genova, M. L. The role of

Coenzyme Q in mitochondrial electron transport. Mitochondrion

2007, 7 (Suppl), S8−33.

(70) Burton, G. W.; Ingold, K. U. beta-Carotene: an unusual type of

lipid antioxidant. Science 1984, 224 (4649), 569−73.

(71) Kennedy, T. A.; Liebler, D. C. Peroxyl radical scavenging by

beta-carotene in lipid bilayers. Effect of oxygen partial pressure. J. Biol.

Chem. 1992, 267 (7), 4658−63.

(72) Palozza, P.; Calviello, G.; Bartoli, G. M. Prooxidant activity of

beta-carotene under 100% oxygen pressure in rat liver microsomes.

Free Radical Biol. Med. 1995, 19 (6), 887−92.

(73) Tsuchihashi, H.; Kigoshi, M.; Iwatsuki, M.; Niki, E. Action of

beta-carotene as an antioxidant against lipid peroxidation. Arch.

Biochem. Biophys. 1995, 323 (1), 137−47.

(53) Kohno, H.; Suzuki, R.; Noguchi, R.; Hosokawa, M.; Miyashita,

K.; Tanaka, T. Dietary conjugated linolenic acid inhibits azoxy-

methane-induced colonic aberrant crypt foci in rats. Jpn. J. Cancer Res.

2002, 93 (2), 133−42.

(54) Moloney, F.; Toomey, S.; Noone, E.; Nugent, A.; Allan, B.;

Loscher, C. E.; Roche, H. M. Antidiabetic effects of cis-9, trans-11-

O

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry

pubs.acs.org/joc

Article

(74) Mordi, R. C.; Walton, J. C.; Burton, G. W.; Hughes, L.; Ingold,

K. U.; Lindsay, D. A. Exploratory study of β -carotene autoxidation.

Tetrahedron Lett. 1991, 32 (33), 4203−4206.

(75) Mordi, R. C.; Walton, J. C.; Burton, G. W.; Hughes, L.; Keith, I.

U.; David, L. A.; Douglas, M. J. Oxidative degradation of β -carotene

and β -apo-8 ′-carotenal. Tetrahedron 1993, 49 (4), 911−928.

(76) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson,

G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.;

Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J.

V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams; Ding, F.; Lipparini,

F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.;

Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.;

Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;

Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;

Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.;

Bearpark, M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov,

V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.;

Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;

Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.;

Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J.

Gaussian 16, Rev. C.01; Gaussian, Inc.: Wallingford, CT, 2016.

P

https://dx.doi.org/10.1021/acs.joc.0c01920

J. Org. Chem. XXXX, XXX, XXX−XXX

Article Doi

Development and Application of a Peroxyl Radical Clock Approach for Measuring Both Hydrogen-Atom Transfer and Peroxyl Radical Addition Rate Constants

DOI: 10.1021/acs.joc.0c01920

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.joc.0c01920

Full text of DOI:10.1021/acs.joc.0c01920

Products guided by the article

R&D Labs maybe for 90760-92-2

Relevant to this article

Hot Product