New Mechanism for Cinchona Alkaloid-Catalysis Allows for an Efficient Thiophosphorylation Reaction

E ﬃ cient Thiophosphorylation Reaction

Article

New Mechanism for Cinchona Alkaloid-Catalysis Allows for an

∇

Nastaran Salehi Marzijarani,* Yu-hong Lam, Xiao Wang, Artis Klapars, Ji Qi, Zhiyan Song,

∇

Benjamin D. Sherry, Zhijian Liu, and Yining Ji*

Cite This: https://dx.doi.org/10.1021/jacs.0c09192

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: An eﬃcient synthesis of nucleoside 5′-monothio-

phosphates under mild reaction conditions using commercially

available thiophosphoryl chloride was achieved with a cinchona

alkaloid catalyst. A detailed mechanistic study of the reaction was

undertaken, employing a combination of reaction kinetics, NMR

spectroscopy, and computational modeling, to better understand

the observed reactivity. Taken collectively, the results support an

unprecedented mechanism for this class of organocatalyst.

INTRODUCTION

solvents and provide low yields. We hypothesized that the

unique properties of cinchona alkaloids could be harnessed to

facilitate a selective thiophosphorylation reaction, achieved by

■

Cinchona alkaloids represent a family of privileged catalysts

that eﬃciently promote many classes of organic reactions in a

highly selective fashion. These naturally occurring alkaloids

contain an array of key functional groups that cooperatively

enforce precise substrate-catalyst interactions in transition

states. The key features responsible for the successful

deployment of cinchona alkaloids in synthetic chemistry are

not only their capability for multisite binding but also their

activation of the nucleoside and/or PSCl to furnish, after

hydrolysis, the corresponding biologically active phosphor-

othioates.

In this paper, we present the development of a novel and

eﬃcient cinchona alkaloid-catalyzed thiophosphorylation of

nucleosides (Scheme 1). Quinine as the catalyst not only

conformational ﬂexibility.

Scheme 1. Direct Thiophosphorylation of Nucleosides

Despite extensive exploration of cinchona alkaloids as

catalysts for inducing stereoselectivity in organic reactions,

their utility in enhancing reactivity and site selectivity has been

less studied. In particular, we became intrigued by the

utilization of cinchona alkaloid catalysts in the context of

nucleoside thiophosphorylation, a reaction of great importance

in the design of new nucleotide analogs as therapeutic agents.

accelerated the reaction but also eliminated the troublesome

side products. The remarkable eﬃciency of the quinine-

catalyzed reaction prompted a detailed analysis of the

mechanism of the title reaction via reaction kinetics,

spectroscopic analysis (NMR), and theoretical (DFT)

approaches. The routinely proposed mechanism for cinchona

alkaloid catalysis invokes the involvement of the quinuclidine

moiety as a base and the secondary alcohol as a hydrogen-bond

Several nucleotides have been identiﬁed as potential

anticancer, antiviral, or antimicrobial agents, not only due to

their nucleoside recognition as a key element but also the

phosphate moiety which serves as an irreplaceable component

that is required for activity. Substitution of the phosphate

oxygen atom by sulfur, known as the thio eﬀect, has led to

improved cellular uptake and necessary resistance against

hydrolytic enzymes to ensure higher metabolic stability. As

such, the eﬃcient synthesis of nucleoside 5′-monothiophos-

Received: August 26, 2020

phates is of paramount importance, although this has been a

challenging endeavor, particularly due to the lengthy reaction

sequences and protecting group manipulations. We were

particularly interested in the direct thiophosphorylation of

nucleosides using commercially available thiophosphoryl

chloride (PSCl₃). Current methods require high-boiling

XXXX American Chemical Society

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

was reversed (Figure 1 a). During our mechanistic inves-

Article

donor, although a recent report postulated a Brønsted acid-

hydrogen bonding model wherein the role of the two groups

(Table 1). To minimize the formation of chlorination side-

product (1f), the optimization was carried out at 0 °C. A clear

preference for the formation of the undesired regioisomer 1c

(

C2′−OH) was observed when KO Bu and DBU bases were

used (Table 1, entries 1 and 2). The regioselectivity in favor of

the desired isomer 1b was increased after we switched to

triethylamine (entries 3); however, 1d and 1e byproducts were

formed competitively. The overall selectivity was improved by

employing a less basic pyridine derivative such as 2,6-lutidine

(Table 1, entry 4), although the reaction conversion remained

Figure 1. (a) Established and (b) newly uncovered mode of

activation of cinchona alkaloids; Nu = nucleophile, E = electrophile.

poor due to low electrophilicity of PSCl in the absence of any

catalysts for activation. Given the low reactivity observed with

the conditions described above, we hypothesized that

tigation, we uncovered a novel quinine activation model

distinct from these prior proposals through the quinoline

nitrogen atom (Figure 1b), while DFT transition state

calculations unraveled a multipoint recognition of the substrate

by the catalyst to stabilize a speciﬁc conformation in the

transition state before the construction of an oxygen

phosphorus bond.

activation of PSCl via a series of nucleophilic catalysts may

lead to improved conversion (see the Supporting Information

for more details). Experimentally, in the presence of 2,6-

lutidine as a proton scavenger, reactivity improved from 7% to

11−12% with the addition of 25 mol % pyridine or 6-

methoxyquinoline (Table 1, entries 5 and 10). In the presence

of more nucleophilic pyridine derivatives such as DMAP, the

conversion increased to 26% with low yield of the desired

product (Table 1, entry 6). We also tested 4-methylpyridine-

N-oxide, 1-methylimidazole, and 3-nitro-1H-1,2,4-triazole as

catalysts and observed an improvement in conversion to 100%,

RESULTS AND DISCUSSION

■

Reaction Optimization. Nucleoside 1a was chosen for our

initial exploration and optimization studies (Table 1). Our

initial investigations aimed to explore the intrinsic selectivity of

the substrate toward PSCl , in which we evaluated a series of

6%, and 83%, respectively (Table 1, entries 7−9).

bases, demonstrated that site-selective thiophosphorylation

was not readily attainable, as a series of byproducts were also

observed. The regioisomer 1c, the overphosphorylated

impurity 1d, the dimerized side-product 1e (resulting from a

reaction between the nucleoside substrate and product prior to

water quench), and chlorinated impurity 1f (arising from S_N2

chlorination of the ﬁnal dichlorothiophosphate product) all

competed with the desired thiophosphorylation pathway

Unfortunately, only 18−45% of the desired 5′-thiophosphory-

lated product 1b was formed, mainly because of competitive

formation of the dimeric impurity 1e. These strong

nucleophilic catalysts served to render the ﬁnal dichlorothio-

phosphate intermediate more electrophilic (prior to water

quench) and correspondingly more susceptible to dimeriza-

tion.

Table 1. Optimization of the Thiophosphorylation Reaction

entry

activator

base

conv. (%)

1b (%)

1c (%)

1d (%)

1e (%)

none

pyridine

KOtBu

DBU

0.5

0.2

0.7

Et N

2,6-lutidine

0.5

0.7

0.8

0.2

0.1

0.7

0.4

0.7

100

DMAP

4-methylpyridine N-oxide

1-methylimidazole

3-nitro-1H-1,2,4-triazole

6-methoxyquinoline

(DHQD) PYR

(DHQ) PHAL

quinidine

quinine

All entries: ratios of impurities, starting material, and product were measured by HPLC chromatography. ND = not detected.

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

We next turned our attention to exploring cinchona alkaloids

in an eﬀort to enhance reactivity as well as selectivity.

Scheme 2. Substrate Scope for Quinine Catalyzed PSCl₃

Incorporation of Nucleosides

Cinchona alkaloid catalysts, such as (DHQD) PYR,

(

DHQ) PHAL, and quinidine were tested in this reaction

and showed 13%, 75%, and 17% of the desired product 1b,

respectively (Table 1, entries 11−13). Remarkably, using a

stoichiometric amount of quinine in the absence of additional

base resulted in formation of the desired product in high

selectivity, as evidenced by a measured 28% yield at 32%

conversion (Figure 2). Encouraged by this result, the system

was further optimized for solvent, temperature, base, and

quinine loading (Figure 2 and see the Supporting Information

for more details). Using a catalytic amount of quinine (0.25

equiv) in combination with 2,6-lutidine (1.5 equiv) at 0 °C in

THF aﬀorded not only a high conversion of 97%, but also high

selectivity for the desired compound 1b (93%) with only 2% of

the dimer 1e and 0.7% of the overphosphorylated impurity 1d

(

Table 1, entry 14). Unlike other nucleophilic catalysts shown

in Table 1 (entries 7−9), a signiﬁcant reduction on the level of

dimer impurity by quinine was indicative of a more speciﬁc

electrophilic activation of PSCl than dichlorothiophosphate

product.

Quinine (25 mol %) was used for all entries. N-benzylquininium

chloride (50 mol %) was used for 2a, 4a, 7a, 8a substrates. Condition

with 1.4 equiv of PSCl and 1.5 equiv of 2,6-lutidine at 0 °C was used.

Condition with 1.4 equiv of PSCl and 1.5 equiv 2,6-lutidine at 10

C was used. Assay yield (AY) was measured by quantitative HPLC

chromatography, H NMR or P NMR spectroscopy.

however, no known examples exist in which it is employed as a

catalyst for enhancing the reactivity and selectivity of a

thiophosphorylation reaction. To better understand the origin

Figure 2. Importance of the catalyst loading and base in the

thiophosphorylation reaction.

of the improved performance, we studied the mechanism of

this transformation, leveraging a suite of tools and techniques

that helped to shed light on this unique reactivity.

Reaction Kinetic Analysis. Because of the low solubility

of iBu-3′F-guanosine (1a) in the reaction solvent (ca. 2.33

mg/mL in THF) and the preference for homogeneous systems

when conducting kinetics studies, the isobutyryl group was

replaced by an octanoyl group to obtain a more soluble analog

Substrate Scope for Quinine Catalyzed Thiophos-

phorylation Reaction. With optimized conditions in hand,

we explored the applicability and synthetic utility of this

method against a range of nucleosides (both purines and

pyrimidines; Scheme 2). Guanosine and thymidine nucleosides

b (92%) and 3b (85%) provided high assay yields (AY) after

day under the original reaction conditions. For all other

a′ (>20 mg/mL in THF; Scheme 3). Reaction progress

substrates, we observed lower reactivities, although selectivities

kinetic analysis by F NMR spectroscopy followed by

variable time normalization revealed a ﬁrst-order dependence

remained high, therefore a greater excess of PSCl and 2,6-

lutidine was used to achieve high conversions up to 4-day time

period. This methodology was also compatible with uracil,

on the concentration of the nucleoside 1a′, PSCl , and quinine

(Figure 3). It also revealed the reaction rate to be independent

of the concentration of 2,6-lutidine. This kinetic information is

consistent with the turnover-limiting step being the P−O bond

purine, and adenine based nucleosides (5b−10b) to generate

the corresponding monothiophosphates in moderate to good

yields (42−87% AY). It is noteworthy that all these

substrates proved sluggish in reactions with PSCl or showed

Scheme 3. Experiments Designed to Extract Reaction

Orders for Thiophosphorylation of 1a′ Catalyzed by Q1

very low conversions in the absence of any catalysts. In order

to understand the role of quinine, a series of mechanistic

studies were performed (next section), and it was found that

N-benzylquininium chloride Q2 was more activating than

quinine. Applying this catalyst to substrates that showed low

performance such as 2b, 4b, 7b, and 8b, resulted in yield

improvements (up to 71% AY).

Quinine is an attractive catalyst because it is commercially

available, inexpensive, stable, recoverable, and most impor-

tantly, it provided the most promising result in comparison

with other conditions (Table 1). Quinine has been utilized

previously in a phosphorylation reaction between a phosphor-

oamidic acid and a functionalized nucleoside as a base;

Journal of the American Chemical Society

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

involving nucleophilic catalysis by the quinoline nitrogen are

Article

the C-9 position of the quinine shifts rapidly from 5.5 to 6.6

ppm upon the addition of PSCl₃(see the Supporting

Information for details). This change in chemical shift was

determined to be the result of a conformational change

triggered by protonation of the quinuclidine moiety by HCl

generated in situ.

Despite the protonation of the free quinuclidine via HCl in

the reaction mixture, a small amount of deprotonated form

may exist in equilibrium and be responsible for the catalysis.

Therefore, we decided to block the nitrogen of the

quinuclidine completely by synthesizing various N-alkylated

quinine catalysts. The reaction was tested with N-benzyl and

N-pentyl alkylated quinine derivatives (Q2 and Q6), and the

Figure 3. Variable time normalization analysis to enable the

determination of the reaction order in [1a′], [PSCl ], and [Q1].

The normalization of the time scale in 1a′, PSCl , and Q1 leads to the

overlay of all reaction proﬁles in a straight line.

results showed that these catalysts were not only eﬀective but

also provided 4- and 3-fold rate enhancements, respectively,

compared to parent quinine catalyst (Figure 4b). These

experimental results completely ruled out the involvement of

the quinuclidine moiety in activation of the substrate via its

formation through a three-component transition state

composed of the nucleoside, the thiophosphorylating agent,

and the organocatalyst.

Role of C-9 Alcohol. As reported in the literature, the C-9

free nitrogen.

alcohol of a cinchona alkaloid can play a role in substrate

Role of Quinoline. To elucidate the role of the quinoline,

we synthesized several quinine derivatives to study the eﬀect of

diﬀerent substitution on the rate of the reaction. First, we

prepared the bis-quaternary ammonium salt of the quinine

catalyst (Figure 4c, Q7) and discovered that catalysis was

completely inhibited, emphasizing the importance of quinoline

nitrogen for a productive reaction. Next, we synthesized

quinine derivatives with varied substituents at the C-6 position

of the quinoline ring to study the inﬂuence of electronic eﬀects

on the reaction performance. We compared the reactivity of

three catalysts in which OMe (R group) was replaced by

ﬂuorine (Q8), hydrogen (Q9), and methyl (Q10) groups

activation through hydrogen bonding. Strikingly, in our

system, functionalizing the C-9 alcohol as either a methoxy

group (Q3) or as the phenylcarbamate (Q4) led to essentially

identical rates as reactions using the parent quinine catalyst

(Q1 ), with k _r _e _l = 1.01 and 1.14, respectively (Figure 4 a). Even

(

Figure 4c). By comparing relative rates, it was evident that the

rate was impacted by the nature of the substituent (R) on the

quinoline moiety. The reaction was faster when catalyzed by a

quinine derivative with more electron-donating groups, the

highest rate k_r_e_lof 1.0 for a methoxy substituent when using

quinine (Q1) and the lowest k_r_e_lof 0.29 for a ﬂuoride group

(

Q8). The observed electronic eﬀect of the substituents on the

rates is consistent with the role of quinoline as a nucleophilic

catalyst for the activation of the electrophile PSCl . In good

agreement with this hypothesis, we observed a positive linear

correlation between HOMO energies and rates of reactions

obtained for diﬀerent quinine derivatives (Q1 and Q8−10),

i.e., the higher the HOMO energy, the faster the rate, lending

further support to our proposal for the engagement of the

quinoline moiety in the nucleophilic catalysis (Figure 5).

Although the nucleophilicity of the two nitrogen atoms has

received occasional attention in the literature, mechanisms

Figure 4. Eﬀects of C-9 substitution (a), N-alkylation of the

quinuclidine moiety (b), diﬀerent substitution on the quinoline ring

(

c), and the chiral pocket (d) on the rate of thiophosphorylation

reaction. All of the k_r_e_lhave been calculated relative to quinine (Q1).

more surprising, the removal of the hydroxyl group (Q5) led to

only a slight decrease in the reaction rate (k_r_e_l= 0.88). These

results collectively argue against an activation pathway

involving the secondary alcohol of the catalyst.

Role of Quinuclidine. In general, the quinuclidine

fragment present in quinine is known to function either as a

base in Brønsted base catalysis or after protonation, as an

activator of electrophiles via Brønsted acid catalysis (Figure

Figure 5. Correlation between DFT (wB97XD/6-31G**)-computed

HOMO energies of quinoline derivatives versus reaction rates for

quinine-catalyzed thiophosphorylation reaction.

a). In our case, by examining the reaction course via H NMR

spectroscopy, it was observed that the oxymethine proton at

Journal of the American Chemical Society

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Figure 6. DFT (wB97XD/6-31G**)-computed transition structures for the reactions of C5′−OH of 1a″ with PSCl activated by quinine Q1 (left)

or Q2 (right). The noncritical hydrogen atoms are omitted for clarity.

Scheme 4. Proposed Catalytic Cycle

much rarer, if at all precedented, than mechanisms involving

responsible for the inferior reactivity when compared to the

model substrate (Scheme 2). These observations indicate a

speciﬁc mode of interaction between the reactants and the

catalyst, in analogy to what is typically observed in an enzyme

active site.

quinuclidine nitrogen. This newly discovered activation

mode is likely to be of interest beyond the current

transformation.

Importance of the Chiral Pocket. Next, we investigated

the eﬀect of the chiral pocket by comparing the rates of

catalysis between quinine, its derivatives, and achiral quinolines

as shown in Figure 4d (see the Supporting Information for

more details). The reaction rate for a mixture of 6-methoxy-4-

methylquinoline and quinuclidine was similar to the back-

ground reaction, k_r_e_lof 0.17 and 0.16, respectively. An

enhanced reactivity (k_r_e_l= 0.31) was observed by tethering

the two functionalities (quinoline and a tertiary amine) in the

form of a new catalyst (Q11). More insightful, relative rate

comparison of quinine (Q1) with its pseudoenantiomer

quinidine (Q12), epimeric at C-8 and C-9, (k_r_e_l= 1.0 vs

DFT Calculations. The experimentally derived rate law

supports the involvement of one nucleoside, one thiophos-

phorylating agent, and one organocatalyst at the rate-

determining transition state (Figure 3). To corroborate the

kinetic ﬁndings and to gain structural insights as to how the

nucleoside substrate ﬁts into the chiral pocket provided by the

organocatalyst, we modeled the transition structures (TSs) for

the reaction of the acetyl-protected analog of 1a (1a″) (Figure

6), 5a and 9a (see the Supporting Information for details).

These TSs correspond to backside concerted displacement at

phosphorus, in which the substrate hydroxyl group acts as the

nucleophile and the catalyst quinoline ring acts as the leaving

group, and the thiophosphorylating agent is activated by

quinine through its quinoline nitrogen. In TS-1a, the

protonated quinuclidine donates a hydrogen bond to

guanosine ring carbonyl oxygen, while the guanosine C2-NH

hydrogen bonds with the chloride anion, which deprotonates

the primary hydroxyl group as the O−P bond is formed.

The geometry of TS-1a is also consistent with the

structure−reactivity correlations with quinine derivatives

(Figure 4). Both the C-9 hydroxyl group and the vinyl group

of quinine point toward solvent space, corroborating the

.19), is consistent with a match-mismatch scenario with

respect to the nucleoside. Similarly, 9-epi-quinine (Q13)

resulted in a reduced rate (k_r_e_l= 0.30). The latter results are

strongly suggestive that the conﬁguration of the chiral pocket is

essential to the productive organization of reaction compo-

nents for the thiophosphorylation reaction, and other stereo-

chemical conﬁgurations diﬀerent from the parent quinine is

less than ideal for reactivity. Similarly, the overall conformation

of the nucleoside is important as well, and should match the

catalyst. The diﬀerences in conformation, induced by varying

the substituents of the sugar ring or the nucleobase are possibly

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

https://pubs.acs.org/doi/10.1021/jacs.0c09192.

Article

negligible eﬀects of these groups on the catalytic performance.

Indeed, dihydroquinine (DHQ, Q14) was experimentally

found to deliver similar catalyst performance compared with

the parent quinine (see the Supporting Information for

details). Likewise, the space around the methoxy group on

quinoline (Figure 4c) is highly accessible and could

accommodate various substituents of diﬀerent electronic

character with minimal steric interference (Figure 6).

paradigm for design and use of these catalysts in the

phosphorylation ﬁeld and beyond.

ASSOCIATED CONTENT

sı Supporting Information

The Supporting Information is available free of charge at

■

All experimental procedures, reaction optimizations,

complete characterization (NMR, MS) for all new

compounds, kinetic data, and xyz coordinates of DFT

computed structures (PDF)

The TS for the Q2-catalyzed reaction was also explored

Figure 6). Even with the quinuclidine nitrogen benzylated, the

(

chiral scaﬀold was found to accommodate the C5′-

thiophosphorylation transition state well. In TS-2a, the

guanosine C6 carbonyl oxygen is engaged in nonclassical

hydrogen bonds with three C−H groups adjacent to the

AUTHOR INFORMATION

Corresponding Authors

■

quaternary ammonium N . Thus, the geometries of TS-1a

and TS-2a shed light on the manner in which the chiral pocket

of quinine complements the shape of the chiral nucleoside in

the thiophosphorylation reaction. The high levels of reactivity

and selectivity experimentally observed can be explained

through multiple noncovalent interactions between the

Nastaran Salehi Marzijarani − Department of Process

Research and Development, Merck & Co., Inc., Rahway, New

Jersey 07065, United States; orcid.org/0000-0001-8940-

4674 ; Email: nastaran.salehi.marzijarani@merck.com

Yining Ji − Department of Process Research and Development,

Merck & Co., Inc., Rahway, New Jersey 07065, United

States; orcid.org/0000-0002-9650-6844;

substrate and the catalyst.

Mechanistic Proposal. Based on the collective evidence

obtained from reaction kinetics, NMR spectroscopy, and

computational studies, a plausible mechanism for the

thiophosphorylation of octanoyl-3′-ﬂuoro-guanosine 1a′

Email: yining.jichen@merck.com

Authors

Yu-hong Lam − Department of Computational and Structural

Chemistry, Merck & Co., Inc., Rahway, New Jersey 07065,

United States; orcid.org/0000-0002-4946-1487

Xiao Wang − Department of Analytical Research and

Development, Merck & Co., Inc., Rahway, New Jersey 07065,

United States; orcid.org/0000-0003-3238-5366

Artis Klapars − Department of Process Research and

Development, Merck & Co., Inc., Rahway, New Jersey 07065,

United States

Ji Qi − Department of Process Research and Development,

Merck & Co., Inc., Rahway, New Jersey 07065, United

States; Department of Process Research and Development,

MSD R&D (China) Co., Ltd., Beijing 100012, China;

orcid.org/0000-0003-4585-6534

using PSCl catalyzed by quinine is proposed (Scheme 4). A

fast equilibrium is established between quinine hydrochloride

^aⁿ^d^P^S^C^l3^. When the two-component complex encounters a

molecule of nucleoside, the desired thiophosphorylation will

occur through the proposed pentavalent transition state to

deliver the desired product 1b′. The catalyst turnover is then

facilitated by 2,6-lutidine, serving as a proton scavenger.

Evidence for a plausible general base catalysis is provided by

the observation of a positive KIE of 1.7 when using substrate

with deuterated hydroxyl groups at both the C5 ′ and C2′

positions of the nucleoside (see the Supporting Information for

details).

CONCLUSIONS

Zhiyan Song − Department of Synthetic Chemistry,

Pharmaron Beijing Co., Ltd., Beijing 100176, China

Benjamin D. Sherry − Department of Process Research and

Development, Merck & Co., Inc., Rahway, New Jersey 07065,

United States

Zhijian Liu − Department of Process Research and

Development, Merck & Co., Inc., Rahway, New Jersey 07065,

United States; orcid.org/0000-0002-5750-9890

■

In conclusion, we have developed a novel and direct quinine-

catalyzed 5′-thiophosphorylation of a series of nucleosides

using commercially available PSCl . Extensive NMR, kinetics,

computational modeling, and catalyst derivatization studies

provided insights into a multipoint recognition by the catalyst

favoring a preferable conformation of the substrate in the

transition state to enable a highly selective thiophosphorylation

reaction. The most alluring feature of our mechanistic ﬁndings

is the impact of relatively weak noncovalent catalyst−substrate

interactions that activated the reaction components simulta-

neously or, in another words, multiple sites were in charge of

ﬁne-tuning the catalytic activity.

We discovered three key elements for the catalytic activity of

quinine: (1) quinine’s chiral pocket governed both reactivity

and selectivity, (2) plausible general base catalysis by the

chloride counterion of the quinuclidinium, and (3) unexpected

Complete contact information is available at:

https://pubs.acs.org/10.1021/jacs.0c09192

Author Contributions

These authors contributed equally.

Notes

∇

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

activation of PSCl by the nitrogen of the quinoline moiety.

We acknowledge Nelo Rivera, Daniel Zewge, Xiaodong Bu,

Fuh-Rong Tsay, Imad Ahmad Haidar, Erik Regalado, Jimmy

DaSilva, and John Kong for analytical support. We greatly

acknowledge Neil Strotman, Eugene E. Kwan, Dan Lehnherr,

Mikhail Reibarkh, Edward Sherer, Alan Hyde, Rebecca T.

Ruck, Louis-Charles Campeau, Paul G. Bulger, and especially

Babak Borhan for helpful discussions. We are grateful for all

Studying the mechanism and key factors that control selectivity

in cinchona alkaloid organocatalysts is a necessary step to

understand their catalytic eﬃciency and broaden their

applications. Considering the long history of cinchona

alkaloids as privileged catalysts, this unique mechanistic path

should be of great interest to chemists, as it opens a new

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

therapy with nucleotide/nucleoside analogs after curative treatment

Article

the support from external capabilities, especially Shiping Ye

and Jake Zhiguo Song.

for patients with hepatitis B virus-related hepatocellular carcinoma: A

systematic review and meta-analysis. Clin. Res. Hepatol. Gastroenterol.

6) (a) Jordheim, L. P.; Durantel, D.; Zoulim, F.; Dumontet, C.

015, 39 (4), 458−468.

REFERENCES

■

(

1) (a) Jacobsen, E. N.; MacMillan, D. W. C. Organocatalysis. Proc.

Advances in the development of nucleoside and nucleotide analogues

Natl. Acad. Sci. U. S. A. 2010, 107 (48), 20618−20619. (b) Miyaji, R.;

Wada, Y.; Matsumoto, A.; Asano, K.; Matsubara, S. Bifunctional

organocatalysts for the asymmetric synthesis of axially chiral

benzamides. Beilstein J. Org. Chem. 2017 , 13 , 1518 −1523. (c) Li,

H.; Wang, Y.; Tang, L.; Deng, L. Highly Enantioselective Conjugate

Addition of Malonate and β -Ketoester to Nitroalkenes: Asymmetric

C −C Bond Formation with New Bifunctional Organic Catalysts

Based on Cinchona Alkaloids. J. Am. Chem. Soc. 2004 , 126 (32),

906 −9907. (d) Taylor, M. S.; Jacobsen, E. N. Asymmetric Catalysis

by Chiral Hydrogen-Bond Donors. Angew. Chem., Int. Ed. 2006, 45

10), 1520−1543. (e) Tian, S.-K.; Chen, Y.; Hang, J.; Tang, L.;

McDaid, P.; Deng, L. Asymmetric Organic Catalysis with Modified

for cancer and viral diseases. Nat. Rev. Drug Discovery 2013 , 12 , 447 −

464. (b) Seley-Radtke, K. L.; Yates, M. K. The evolution of nucleoside

analogue antivirals: A review for chemists and non-chemists. Part 1:

Early structural modifications to the nucleoside scaffold. Antiviral Res.

018, 154, 66−86. (c) Eyer, L.; Nencka, R.; de Clercq, E.; Seley-

Radtke, K.; Ruzek, D. Nucleoside analogs as a rich source of antiviral

agents active against arthropod-borne flaviviruses. Antivir. Chem.

Chemother. 2018, 26 , 1 −28. (d) Parker, L. L.; Schilling, A. B.; Kron, S.

J.; Kent, S. B. H. Optimizing Thiophosphorylation in the Presence of

Competing Phosphorylation with MALDI-TOF −MS Detection. J.

Proteome Res. 2005 , 4 (5), 1863 −1866. (e) Murray, A. W.; Atkinson,

M. R. Adenosine 5 ′-phosphorothioate. A nucleotide analog that is a

substrate, competitive inhibitor, or regulator of some enzymes that

interact with adenosine 5 ′-phosphate. Biochemistry 1968, 7 (11),

(

Cinchona Alkaloids. Acc. Chem. Res. 2004 , 37 (8), 621 −631.

f) Kacprzak, K.; Gawronski, J. Cinchona Alkaloids and Their

Derivatives: Versatile Catalysts and Ligands in Asymmetric Synthesis.

Synthesis 2001 , 07 , 961 −998. (g) Bolm, C.; Schiffers, I.; Dinter, C. L.;

Gerlach, A. Practical and Highly Enantioselective Ring Opening of

Cyclic Meso-Anhydrides Mediated by Cinchona Alkaloids. J. Org.

Chem. 2000 , 65 (21), 6984 −6991. (h) Marcelli, T.; Hiemstra, H.

Cinchona Alkaloids in Asymmetric Organocatalysis. Synthesis 2010,

4023−4029. (f) Vosberg, H. P.; Eckstein, F. Effect of deoxynucleoside

phosphorothioates incorporated in DNA on cleavage by restriction

enzymes. J. Biol. Chem. 1982 , 257 (11), 6595 −6599. (g) Wickstrom,

E. Oligodeoxynucleotide stability in subcellular extracts and culture

media. J. Biochem. Biophys. Methods 1986, 13 (2), 97−102.

(h) Kowalska, J.; Lukaszewicz, M.; Zuberek, J.; Ziemniak, M.;

8 , 1229 −1279. (i) Vakulya, B.; Varga, S.; Csampai, A.; Soos, T.

́ ́

Highly Enantioselective Conjugate Addition of Nitromethane to

Chalcones Using Bifunctional Cinchona Organocatalysts. Org. Lett.

Darzynkiewicz, E.; Jemielity, J. Phosphorothioate analogs of m7GTP

are enzymatically stable inhibitors of cap-dependent translation.

Bioorg. Med. Chem. Lett. 2009, 19 (7), 1921−1925.

005 , 7 (10), 1967 −1969. (j) Doyle, A. G.; Jacobsen, E. N. Small-

Molecule H-Bond Donors in Asymmetric Catalysis. Chem. Rev. 2007 ,

07 (12), 5713 −5743. (k) Yao, Y.; Xu, L. Organocatalytic

Applications of Cinchona Alkaloids as a Potent Hydrogen-bond

Donor and/or Acceptor to Asymmetric Transformations: A Review.

Mini-Rev. Org. Chem. 2016 , 13 , 184 −197. (l) Connon, S. J.

Asymmetric catalysis with bifunctional cinchona alkaloid-based urea

and thiourea organocatalysts. Chem. Commun. (Cambridge, U. K.)

(

7) (a) Pradere, U.; Garnier-Amblard, E. C.; Coats, S. J.; Amblard,

F.; Schinazi, R. F. Synthesis of Nucleoside Phosphate and

Phosphonate Prodrugs. Chem. Rev. 2014, 114 (18), 9154−9218.

(b) Knouse, K. W.; deGruyter, J. N.; Schmidt, M. A.; Zheng, B.;

Vantourout, J. C.; Kingston, C.; Mercer, S. E.; McDonald, I. M.;

Olson, R. E.; Zhu, Y.; Hang, C.; Zhu, J.; Yuan, C.; Wang, Q.; Park, P.;

Eastgate, M. D.; Baran, P. S. Unlocking P(V): Reagents for chiral

phosphorothioate synthesis. Science 2018 , 361 (6408), 1234 −1238.

008, 22, 2499−2510. (m) Song, C. E. Cinchona Alkaloids. In

8) (a) Cook, A. F. Nucleoside S-alkyl phosphorothioates. IV.

Synthesis and Catalysis; Song, C. E., Ed.; Wiley-VCH: Weinheim,

Germany, 2009; pp 1−10.

Synthesis of nucleoside phosphorothioate monoesters. J. Am. Chem.

Soc. 1970, 92 (1), 190 −195. (b) Szczepanik, M. B.; Desaubry, L.;

Johnson, R. A. Synthesis of Nucleoside 3 ′-Thiophosphates in One

Step Procedure AU - Szczepanik, Maciej B. Nucleosides Nucleotides

(

2) (a) Dijkstra, G. D. H.; Kellogg, R. M.; Wynberg, H.; Svendsen, J.

S.; Marko, I.; Sharpless, K. B. Conformational study of cinchona

alkaloids. A combined NMR, molecular mechanics and x-ray

approach. J. Am. Chem. Soc. 1989, 111 (21), 8069−8076.

1999 , 18 (4 −5), 951 −953. (c) Saba, D.; Dekker, C. A. Synthesis of

nucleoside 3 ′-(S-alkyl phosphorothioates) and their use as substrates

for nucleases. Biochemistry 1981 , 20 (19), 5461 −5466. (d) Dunaway-

Mariano, D. Nucleoside phosphorothioates I: Thiophosphorylation

of nucleosides with phosphorus thioxy-anions. Tetrahedron 1976 , 32

b) Molchanov, S.; Rowicki, T.; Gryff-Keller, A.; Kozminski, W.

Conformational Equilibrium of Cinchonidine in C6D12 Solution.

Alternative NMR/DFT Approach. J. Phys. Chem. A 2018 , 122 (39),

832 −7841. (c) Burgi, T.; Baiker, A. Conformational Behavior of

Cinchonidine in Different Solvents: A Combined NMR and ab Initio

Investigation. J. Am. Chem. Soc. 1998, 120 (49), 12920 −12926.

d) Silva, T. H. A.; Oliveira, A. B.; De Almeida, W. B. Conformational

(

24), 2991−2996. (e) Eckstein, F. Nucleoside phosphorothioates. J.

Am. Chem. Soc. 1970, 92 (15), 4718−4723. (f) Olesiak, M.;

Krajewska, D.; Wasilewska, E.; Korczynski, D.; Baraniak, J.;

Okruszek, A.; Stec, W. Thiophosphorylation of Biologically Relevant

Alcohols by the Oxathiaphospholane Approach. Synlett 2002, 6 , 967 −

971. (g) Iyer, R. P.; Egan, W.; Regan, J. B.; Beaucage, S. L. 3H-1,2-

Benzodithiole-3-one 1,1-dioxide as an improved sulfurizing reagent in

the solid-phase synthesis of oligodeoxyribonucleoside phosphoro-

thioates. J. Am. Chem. Soc. 1990 , 112 (3), 1253 −1254. (h) Ahmadi-

beni, Y.; Parang, K. Polymer-Bound Oxathiaphospholane: A Solid-

Phase Reagent for Regioselective Monothiophosphorylation and

Monophosphorylation of Unprotected Nucleosides and Carbohy-

analysis of the antimalarial agent quinine. Struct. Chem. 1997, 8 (2),

5−107. (e) Chung, W.-j.; Denmark, S. E. Bifunctional and

Synergistic Catalysis: Lewis Acid Catalysis and Lewis Base-Assisted

Bond Polarization (n → σ*). In Lewis Base Catalysis in Organic

Synthesis; Vedejs, E., Denmark, S. E., Ed.; 2016; pp 1213 −1258.

(

3) Renzi, P.; Bella, M. Non-asymmetric organocatalysis. Chem.

Commun. 2012, 48 (55), 6881 −6896.

(4) (a) Alvarez-Salas, L. M. Nucleic acids as therapeutic agents. Curr.

Top. Med. Chem. 2008, 8 (15), 1379−1404. (b) Xavier, N. M. Recent

Advances on Nucleotide Analogs and Mimetics: Synthesis and

Biological Properties. Reference Module in Chemistry, Molecular Sciences

and Chemical Engineering; Elsevier: Amsterdam, 2017.

drates. Org. Lett. 2005, 7 (10), 1955 −1958. (i) Zielins

ki, W. S.;

Lesnikowski, Z. J.; Stec, W. J. New approach to the synthesis of

nucleoside phosphorothioates. J. Chem. Soc., Chem. Commun. 1976,

19, 772−773. (j) Chen, Z.; Yang, Y.; Zhang, L.; Zhang, L. One-pot

synthesis of fluoro-substituted nicotinamide nucleoside phosphates.

Chin. J. Med. Chem. 2012, 22 (2), 104−112.

(5) (a) Wigler, P. W.; Lozzio, C. B. Cytotoxic nucleotide analogs as

potential anticancer agents. 5-Bromodeoxyuridine 5 ′-methylphosph-

onate. J. Med. Chem. 1972, 15 (10), 1020 −1024. (b) Jordheim, L. P.;

Durantel, D.; Zoulim, F.; Dumontet, C. Advances in the development

of nucleoside and nucleotide analogues for cancer and viral diseases.

Nat. Rev. Drug Discovery 2013, 12 (6), 447−464. (c) Xia, B. W.;

Zhang, Y. C.; Wang, J.; Ding, F. H.; He, X. D. Efficacy of antiviral

(9) (a) Conway, L. P.; Delley, R. J.; Neville, J.; Freeman, G. R.;

Maple, H. J.; Chan, V.; Hall, A. J.; Hodgson, D. R. W. The aqueous N-

phosphorylation and N-thiophosphorylation of aminonucleosides.

RSC Adv. 2014, 4 (73), 38663−38671. (b) Dahl, O. Preparation of

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Nucleoside Phosphorothioates, Phosphorodithioates and Related

Article

Compounds. Sulfur Rep. 1991 , 11 (1), 167 −188. (c) Harger, M. J.

P. A new mechanism for nucleophilic substitution at a thiophosphoryl

(12) Two adenine-based nucleosides 9a and 10a showed inferior

selectivities than the original substrate 1a, mainly due to the

formation of the bis-thiophosphate and dimer impurities.

centre revealed by the reaction of diisopropylamine with PSCl . Chem.

(13) Some of the challenges associated with these products are high

polarity and diﬃculty of puriﬁcation by reversed-phase chromatog-

raphy, leading to lower isolation yields than assay yields.

(14) Tran, K.; Beutner, G. L.; Schmidt, M.; Janey, J.; Chen, K.;

Rosso, V.; Eastgate, M. D. Development of a Diastereoselective

Phosphorylation of a Complex Nucleoside via Dynamic Kinetic

Resolution. J. Org. Chem. 2015 , 80 (10), 4994 −5003.

(15) (a) Blackmond, D. G. Reaction Progress Kinetic Analysis: A

Powerful Methodology for Mechanistic Studies of Complex Catalytic

Reactions. Angew. Chem., Int. Ed. 2005 , 44 (28), 4302 −4320.

Commun. 2005, No. 22, 2863−2865.

(10) (a) Chladek, S.; Nagyvary, J. Nucleophilic reactions of some

nucleoside phosphorothioates. J. Am. Chem. Soc. 1972 , 94 (6), 2079 −

2085. (b) Moran, J. R.; Whitesides, G. M. A practical enzymic

synthesis of (Sp)-adenosine 5 ′-O-(1-thiotriphosphate) ((Sp)-ATP-.al-

pha.-S). J. Org. Chem. 1984, 49 (4), 704−706.

(11) (a) Cucinotta, C. S.; Kosa, M.; Melchiorre, P.; Cavalli, A.;

Gervasio, F. L. Bifunctional Catalysis by Natural Cinchona Alkaloids:

A Mechanism Explained. Chem. - Eur. J. 2009 , 15 (32), 7913 −7921.

(b) Hiemstra, H.; Wynberg, H. Addition of aromatic thiols to

b) Blackmond, D. G. Kinetic Profiling of Catalytic Organic Reactions

as a Mechanistic Tool. J. Am. Chem. Soc. 2015, 137 (34), 10852−

0866.

16) Bures

conjugated cycloalkenones, catalyzed by chiral.beta.-hydroxy amines.

A mechanistic study of homogeneous catalytic asymmetric synthesis.

J. Am. Chem. Soc. 1981 , 103 (2), 417 −430. (c) Grayson, M. N.; Houk,

K. N. Cinchona Alkaloid-Catalyzed Asymmetric Conjugate Additions:

The Bifunctional Brønsted Acid −Hydrogen Bonding Model. J. Am.

Chem. Soc. 2016, 138 (4), 1170 −1173. (d) Tanriver, G.; Dedeoglu,

B.; Catak, S.; Aviyente, V. Computational Studies on Cinchona

Alkaloid-Catalyzed Asymmetric Organic Reactions. Acc. Chem. Res.

, J. Variable Time Normalization Analysis: General

Graphical Elucidation of Reaction Orders from Concentration

Profiles. Angew. Chem., Int. Ed. 2016, 55 (52), 16084−16087.

(17) Olsen, R. A.; Borchardt, D.; Mink, L.; Agarwal, A.; Mueller, L.

J.; Zaera, F. Effect of Protonation on the Conformation of

Cinchonidine. J. Am. Chem. Soc. 2006, 128 (49), 15594−15595.

016 , 49 (6), 1250−1262. (e) Marcelli, T. Organocatalysis: Cinchona

(18) The related quaternized ammonium salts of cinchona alkaloids

catalysts. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2011 , 1 (1), 142 −

have proved to catalyze numerous organic reactions under phase-

transfer conditions.

52. (f) Chauhan, P.; Chimni, S. S. Aromatic hydroxyl group a

hydrogen bonding activator in bifunctional asymmetric organo-

catalysis. RSC Adv. 2012 , 2 (3), 737 −758. (g) Li, H.; Hong, X.

Computational studies of cinchona alkaloid-catalyzed asymmetric

Michael additions. Chin. Chem. Lett. 2018, 29 (11), 1585 −1590.

(19) Additionally, since the use of the alkylated catalyst provided

even faster reaction rates, it is less likely that PSCl is activated via

electrophilic catalysis through the protonated quinuclidinium (Figure

(

a).

20) To explore whether a general base catalysis by the quinoline

(h) Hintermann, L.; Ackerstaff, J.; Boeck, F. Inner Workings of a

Cinchona Alkaloid Catalyzed Oxa-Michael Cyclization: Evidence for a

Concerted Hydrogen-Bond-Network Mechanism. Chem. - Eur. J.

nitrogen could be operative, relative basicities were obtained for

various 6-substituted model quinolines (see the Supporting

Information for details). The DFT calculations indicated a decreasing

basicity in the order of R = Me > R = OMe > R = H > R = F. The

methyl substituent would render the most basic quinoline nitrogen

atom and fastest reaction if a base role is operative. However, this is

opposite to our experimental observations as shown in Figure 4 c.

2013, 19 (7), 2311 −2321. (i) Zhu, J.-L.; Zhang, Y.; Liu, C.; Zheng,

A.-M.; Wang, W. Insights into the Dual Activation Mechanism

Involving Bifunctional Cinchona Alkaloid Thiourea Organocatalysts:

An NMR and DFT Study. J. Org. Chem. 2012, 77 (21), 9813−9825.

(j) Xue, X.-S.; Li, X.; Yu, A.; Yang, C.; Song, C.; Cheng, J.-P.

Mechanism and Selectivity of Bioinspired Cinchona Alkaloid

Derivatives Catalyzed Asymmetric Olefin Isomerization: A Computa-

tional Study. J. Am. Chem. Soc. 2013, 135 (20), 7462 −7473.

(21) Baidya, M.; Horn, M.; Zipse, H.; Mayr, H. Organocatalytic

Activity of Cinchona Alkaloids: Which Nitrogen Is More

Nucleophilic? J. Org. Chem. 2009, 74 (18), 7157−7164.

k) Kumpulainen, T.; Qian, J.; Brouwer, A. M. Spectroscopic Study

of a Cinchona Alkaloid-Catalyzed Henry Reaction. ACS Omega 2018,

(2), 1871 −1880. (l) Hammar, P.; Marcelli, T.; Hiemstra, H.; Himo,

(22) (a) Taha, H. A.; Richards, M. R.; Lowary, T. L. Conformational

Analysis of Furanoside-Containing Mono- and Oligosaccharides.

Chem. Rev. 2013 , 113 (3), 1851 −1876. (b) Gaweda, K.; Plazinski,

W. The systematic influence of solvent on the conformational features

of furanosides. Org. Biomol. Chem. 2019, 17 (9), 2479 −2485. (c) Lyu,

S.; Beiranvand, N.; Freindorf, M.; Kraka, E. Interplay of Ring

Puckering and Hydrogen Bonding in Deoxyribonucleosides. J. Phys.

Chem. A 2019, 123 (32), 7087−7103.

F. Density Functional Theory Study of the Cinchona Thiourea-

Catalyzed Henry Reaction: Mechanism and Enantioselectivity. Adv.

Synth. Catal. 2007, 349 (17−18), 2537 −2548. (m) Moran, A.;

Hamilton, A.; Bo, C.; Melchiorre, P. A Mechanistic Rationale for the

9-Amino(9-deoxy)epi Cinchona Alkaloids Catalyzed Asymmetric

Reactions via Iminium Ion Activation of Enones. J. Am. Chem. Soc.

(23) The stoichiometrically complete thiophosphorylation activated

013 , 135 (24), 9091 −9098. (n) de Freitas Martins, E.; Pliego, J. R.

complex consists of the transition structures shown in Figure 6 and a

second chloride that does not participate in any bond-making or

Unraveling the Mechanism of the Cinchoninium Ion Asymmetric

Phase-Transfer-Catalyzed Alkylation Reaction. ACS Catal. 2013 , 3

breaking events. This spectator chloride has been omitted to simplify

(4), 613 −616. (o) Martins, E. F.; Pliego, J. R. A comprehensive

conformational searches and the TSs in this work have one unit of

positive charge.

theoretical investigation of the transition states and a proposed kinetic

model for the cinchoninium ion asymmetric phase-transfer catalyzed

alkylation reaction. J. Mol. Catal. A: Chem. 2016, 417, 192−199.

24) Johnston, R. C.; Cheong, P. H.-Y. C −H ···O Non-Classical

Hydrogen Bonding in the Stereomechanics of Organic Trans-

(p) Lam, Y.-h.; Grayson, M. N.; Holland, M. C.; Simon, A.; Houk, K.

formations: Theory and Recognition. Org. Biomol. Chem. 2013, 11

N. Theory and Modeling of Asymmetric Catalytic Reactions. Acc.

Chem. Res. 2016, 49 (4), 750 −762. (q) Kolb, H. C.; VanNieuwenhze,

M. S.; Sharpless, K. B. Catalytic Asymmetric Dihydroxylation. Chem.

Rev. 1994 , 94 (8), 2483 −2547. (r) Yang, H.; Wong, M. W. Oxyanion

Hole Stabilization by C −H ···O Interaction in a Transition State A

Three-Point Interaction Model for Cinchona Alkaloid-Catalyzed

Asymmetric Methanolysis of meso-Cyclic Anhydrides. J. Am. Chem.

Soc. 2013, 135 (15), 5808−5818. (s) He, C. Q.; Simon, A.; Lam, Y.-

h.; Brunskill, A. P. J.; Yasuda, N.; Tan, J.; Hyde, A. M.; Sherer, E. C.;

Houk, K. N. Model for the Enantioselectivity of Asymmetric

Intramolecular Alkylations by Bis-Quaternized Cinchona Alkaloid-

Derived Catalysts. J. Org. Chem. 2017, 82, 8645−8650.

(

31), 5057−5064.

(

25) To compare the performance of Q1 and Q2 catalysts in terms

of activation barrier, we would need not only the two lowest-energy

transition structures, but also the global minimum for the two catalyst·

substrate complexes. The number of complexes that need to be

individually calculated by DFT is intractable, considering the large

molecular size, high conformational ﬂexibility, and the involvement of

ion pairs. Further compounding the problem is the small magnitude

of energy diﬀerence, the 4-fold rate enhancement of Q2 is equivalent

to a reduction in activation barrier by about 0.8 kcal/mol compared to

Q1. Therefore, a direct comparison of Q1 and Q2 to uncover the

J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society

Article

structural origin of the rate diﬀerence between the two catalysts is

highly challenging.

(26) Despite we are proposing the quinine hydrochloride as the

active catalyst, a clear induction period is not observed. As shown in

the Supporting Information , the background reaction might have

played a role in masking the induction period. Besides, as shown by

NMR data, all of the quinine appears to be in the protonated form

after only 10 min of reaction, a relatively short time compared to 22 h

to reach full conversion.