Synthesis of 2-oxoglutarate derivatives and their evaluation as cosubstrates and inhibitors of human aspartate/asparagine-β-hydroxylase

Chemical

Science

View Article Online

View Journal | View Issue

EDGE ARTICLE

Synthesis of 2-oxoglutarate derivatives and their

evaluation as cosubstrates and inhibitors of human

aspartate/asparagine-b-hydroxylase†

Cite this: Chem. Sci., 2021, 12, 1327

All publication charges for this article

have been paid for by the Royal Society

of Chemistry

*

Lennart Brewitz,

Yu Nakashima and Christopher J. Schoﬁeld

2-Oxoglutarate (2OG) is involved in biological processes including oxidations catalyzed by 2OG

oxygenases for which it is a cosubstrate. Eukaryotic 2OG oxygenases have roles in collagen

biosynthesis, lipid metabolism, DNA/RNA modiﬁcation, transcriptional regulation, and the hypoxic

response. Aspartate/asparagine-b-hydroxylase (AspH) is a human 2OG oxygenase catalyzing post-

translational hydroxylation of Asp/Asn-residues in epidermal growth factor-like domains (EGFDs) in

the endoplasmic reticulum. AspH is of chemical interest, because its Fe(^II) cofactor is complexed by

two rather than the typical three residues. AspH is upregulated in hypoxia and is a prognostic marker

on the surface of cancer cells. We describe studies on how derivatives of its natural 2OG cosubstrate

modulate AspH activity. An eﬃcient synthesis of C3- and/or C4-substituted 2OG derivatives,

proceeding via cyanosulfur ylid intermediates, is reported. Mass spectrometry-based AspH assays

with >30 2OG derivatives reveal that some eﬃciently inhibit AspH via competing with 2OG as

evidenced by crystallographic and solution analyses. Other 2OG derivatives can substitute for 2OG

enabling substrate hydroxylation. The results show that subtle changes, e.g. methyl- to ethyl-

substitution, can signiﬁcantly alter the balance between catalysis and inhibition. 3-Methyl-2OG,

a

natural product present in human nutrition, was the most eﬃcient alternative cosubstrate

Received 4th August 2020

Accepted 26th November 2020

identiﬁed; crystallographic analyses reveal the binding mode of (R)-3-methyl-2OG and other 2OG

derivatives to AspH and inform on the balance between turnover and inhibition. The results will

enable the use of 2OG derivatives as mechanistic probes for other 2OG utilizing enzymes and

suggest 2-oxoacids other than 2OG may be employed by some 2OG oxygenases in vivo.

DOI: 10.1039/d0sc04301j

rsc.li/chemical-science

a (co-)substrate, other than the 2OG dehydrogenase complex.

Such enzymes include aminotransferases (e.g. branched-chain

aminotransferases, BCATs),⁵which convert 2OG to glutamate,

Introduction

2-Oxoglutarate (2OG, a-ketoglutarate; 1, Fig. 1a) is an integral

metabolite in most of biology including prokaryotes, archaea,

and eukaryotes;¹2OG is crucially involved in cellular energy

homeostasis and small-molecule metabolism¹and can act as

a signaling molecule linking nitrogen and carbon metabolism.²

2OG is an intermediate of the tricarboxylic acid (TCA) cycle

where it is produced by isocitrate dehydrogenase (IDH)-

catalyzed decarboxylation of ^D-isocitrate; 2OG is converted to

succinyl-CoA and CO₂by the 2OG dehydrogenase complex.

Reductions in cellular 2OG coupled with the formation of (R)-2-

hydroxyglutarate, which occur as a result of IDH mutations,³are

associated with changes in epigenetic regulation (e.g. DNA and

histone methylation status) and certain types of cancer.^3b,4

These eﬀects are proposed to be mediated, at least in part, by

modulation of the activities of enzymes that rely on 2OG as

and 2OG dependent oxygenases.⁶The latter couple the conver-

sion of 2OG and O₂to succinate and CO₂with substrate

hydroxylation or demethylation via hydroxylation.⁷There are

approximately 60–70 assigned human 2OG oxygenases, all

studied members of which likely employ Fe(^II) as a cofactor and

which have diverse roles, ranging from DNA/RNA modication

and damage repair,⁸histone/chromatin modication,⁹lipid

metabolism,¹⁰post-translational modication of proteins with

important functions in the extracellular matrix,^7,11to hypoxia

sensing.¹²

The human 2OG oxygenase aspartate/asparagine-b-

hydroxylase (AspH, BAH, HAAH)¹³is highly unusual amongst

human 2OG dependent hydroxylases, because its Fe(^II)

cofactor is complexed by only two residues (His679 and

His725) rather than by the typical triad of ligands (HXD/E/H)

found in other human 2OG oxygenases (Fig. 1b).^7,14AspH

catalyzes the post-translational hydroxylation of specic Asp-

and Asn-residues in epidermal growth factor-like domains

(EGFDs) of its substrate proteins in the endoplasmic

Chemistry Research Laboratory, University of Oxford, 12 Manseld Road, OX1 3TA,

Oxford, UK. E-mail: christopher.schoeld@chem.ox.ac.uk

† Electronic supplementary information (ESI) available. See DOI:

10.1039/d0sc04301j

Chem. Sci., 2021, 12, 1327–1342 | 1327

View Article Online

Chemical Science

Edge Article

Fig. 1 The AspH active site and stoichiometry of its reaction. (a) AspH catalyzes the post-translational hydroxylation of Asn- and Asp-residues in

epidermal growth factor (EGF)-like domains; (b) analysis of an AspH:substrate (human Factor X, hFX) crystal structure (PDB ID: 5JQY)²²reveals

that two AspH residues (His679 and His725) coordinate the active site metal rather than the typical three residues found in other human 2OG

dependent hydroxylases. In the crystallographic analyses, Mn substitutes for Fe and NOG (3) for 2OG (1); (c) N-oxalylglycine (NOG, 3).

reticulum (Fig. 1a).¹⁵AspH is of signicant interest from 5–6) rather than the well-characterized canonical disulde

a cancer research perspective, because its levels are upregu- connectivity (i.e. Cys 1–3, 2–4, 5–6; ESI Fig. S1†).²²High-

lated in invasive cancers (e.g. hepatocellular carcinoma¹⁶and throughput MS assays were established to monitor the cata-

pancreatic cancer¹⁷) and it is translocated to the cell surface lytic activity of AspH using stable thioether-linked cyclic peptide

where it can be used as a diagnostic and prognostic marker.¹⁸substrate analogues mimicking the central non-canonical

Mouse models¹⁹and heritable genetic diseases associated macrocyclic Cys 3–4 EGFD disulde (ESI Fig. S1†).²³Kinetic

with mutations likely eﬀecting AspH catalysis (e.g. Traboulsi studies have revealed that AspH is sensitive towards subtle

syndrome)²⁰suggest that the Notch signaling pathway may be changes in oxygen availability and thus it is a candidate oxy-

involved in transmitting the eﬀect of AspH on cancer inva- genase for involvement in hypoxia sensing.²³

siveness. AspH levels are regulated by hypoxia which is

Previous studies have revealed that diﬀerences in the

a characteristic of many tumor cells.²¹Thus, AspH appears to cosubstrate binding sites of 2OG oxygenases can be exploi-

be an attractive medicinal chemistry and diagnostic target for ted for stereoselective selective inhibition employing 2OG

certain types of cancer.

analogues, e.g. N-oxalylamino acids.²⁴Analysis of reported

Recent studies have provided crystallographic and solution- AspH crystal structures wherein 2OG is replaced by a close

based evidence that AspH accepts EGFD substrates with an 2OG analogue, i.e. N-oxalylglycine (NOG, 3; Fig. 1c),²²

unusual non-canonical disulde connectivity (i.e. Cys 1–2, 3–4, suggest that the AspH active site is suﬃciently spacious to

Fig. 2 Selected reported strategies for the synthesis of C3- and/or C4-substituted 2OG derivatives. (a) Alkylation reactions,²⁵(b) Michael

reactions,²⁷and (b and c) oxidation reactions^28a–c,29have been employed in syntheses of C3/C4-substituted 2OG derivatives (6).

1328 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

accommodate substituents at the 2OG C3- and/or C4- derivatives can bind at the AspH active site and that subtle

position, in part due to its unusual Fe(^II)-binding geometry diﬀerences in the 2OG substitution pattern can cause signi-

(Fig. 1b). We were therefore interested to test this by cant disturbances in the balance between productive catalysis

exploring how a diverse set of 2OG derivatives interact with and inhibition.

AspH.

To our knowledge, only a limited number of studies

describing how 2OG derivatives bearing substituents at the C3-

Results

and/or C4-position modulate the activities of 2OG oxygenases

(i.e. human JmjC histone N³-methyl lysine demethylase 4A,

KDM4A;²⁵human factor inhibiting the hypoxia-inducible tran-

scription factor HIF-a, FIH;²⁴and rat g-butyrobetaine dioxyge-

nase, BBOX²⁶) are reported. In part, this likely reects a lack of

a simple synthetic method to access these types of 2OG deriv-

atives. Some prior syntheses of C3/C4-substituted 2OG deriva-

tives have relied inter alia on alkylation (Fig. 2a)²⁵and Michael

reactions (Fig. 2b)²⁷to access key synthetic intermediates; the

corresponding 2OG derivatives (6) were obtained aer saponi-

cation. Other approaches rely on oxidation reactions using

ozone²⁸or sodium periodate²⁹as oxidants to convert Michael

acceptors (10) into 2OG derivatives (Fig. 2c). The described

syntheses are frequently associated with limited scalability, low

overall chemical yields, and/or narrow substrate scopes due to

harsh reaction conditions requiring, for example, the use of

strong bases and acids,^25,30high pressure,²⁷or strong

oxidants.^28a–c,29

Cyanophosphorous ylids are reported as valuable interme-

diates for the synthesis of a-keto acids,³¹however, only one

example of a cyanophosphorous ylid (i.e. 11) being converted

into a 2OG derivative is reported (Fig. 2c),³²possibly reecting

limitations associated with the conversion of cyanophos-

phorous ylids into a-keto acids which requires strong

oxidants.^31–33The use of cyanophosphorous ylids has been

largely superseded by the corresponding cyanosulfur ylids,

which can be oxidized under milder conditions.^33a,34We thus

envisaged that cyanosulfur ylids could be used for the synthesis

of C3/C4-substituted 2OG derivatives.

Synthesis of 2OG derivatives

A versatile and scalable synthetic route to access 2OG deriva-

tives was developed employing cyanosulfur ylids as key inter-

mediates (Scheme 1). The route employs mono-methyl

dicarboxylic acid half-esters 13 as starting materials which were

either commercially available or synthesized by established

reactions, i.e. nucleophilic openings of the requisite symmetric

cyclic anhydrides, formylation reactions of aryl iodides,³⁵Heck

couplings³⁶of aryl iodides with orthogonally protected itaco-

nates,³⁷or Horner–Wadsworth–Emmons (HWE)³⁸reactions (ESI

Fig. S2†). In our work, racemic mixtures of mono-methyl

dicarboxylic acid half-esters 13 bearing stereogenic carbon

atoms at the 2OG C3- or C4-equivalent position were employed.

Cyanosulfur ylids 15, which are the key intermediates in

our strategy, were obtained by reaction of mono-methyl

dicarboxylic acid half-esters 13 with the reported tetrahy-

drothiophene bromide salt 14^33ain yields ranging from 11 to

95%. T3P³⁹was chosen as the coupling reagent because it is

suited for use with sterically hindered carboxylic acids,

including those bearing substituents at the carboxylate a-

position. For some substrates, T3P-derived byproducts inter-

fered with the purication process; however, these were

completely removed aer the subsequent reaction by

chromatography.

The cyanosulfur ylids 15 were converted into the corre-

sponding dimethyl dicarboxylic acid esters 12 using oxone^33aas

a mild oxidation reagent in methanol, in part to avoid ester

exchange (Scheme 1). The dimethyl esters 12 were obtained in

high purity aer column chromatography in yields ranging

from 39 to 98%. Lithium hydroxide-mediated saponication of

dimethyl dicarboxylic acid esters 12 aﬀorded the desired 2OG

derivatives 6 (Scheme 1). The 2OG derivatives were obtained in

suﬃcient purity aer removal of excess base by acidic ion

exchange chromatography yielding salt-free dicarboxylic acids 6

(ESI†), which are suitable for performing in vitro biochemical

experiments with AspH. The 2OG derivatives and their synthetic

precursors were stable when stored at ꢀ20 ^ꢁC for more than six

months.

Here we report the use of cyanosulfur ylids as intermediates

that enable the facile synthesis of multiple 2OG derivatives

bearing a diverse set of substituents at the C3- and/or C4-

positions. The synthetic 2OG derivatives were used to modu-

late the activity of recombinant human AspH. Kinetic and

crystallographic studies were employed to elucidate the mech-

anisms by which the 2OG derivatives modulate AspH activity

and to garner information of the active site requirements of

AspH. The results reveal an unexpectedly diverse set of 2OG

Scheme 1 Route and substrate scope for the synthesis of 2OG derivatives. Reagents and conditions: (a) T3P, ⁱPr₂NEt, CH₂Cl₂, 0 ^ꢁC to rt, 11–95%;

(b) oxone, MeOH/H₂O, rt, 39–98%; (c) LiOH, MeOH/H₂O, 0 ^ꢁC to rt, then: puriﬁcation by ion exchange chromatography (Dowex® 50XW8), 67%

to apparent quantitative.

Chem. Sci., 2021, 12, 1327–1342 | 1329

View Article Online

Chemical Science

Edge Article

substituents at the 2OG C3- and/or C4-positions (Table 1,

entries 1–21). The aliphatic substituents varied in both length

and steric bulk of the carbon chain. Furthermore, 2OG deriva-

tives were synthesized in which the C3/C4 ethylene unit of 2OG

Scope of the synthesis

Following its development, the synthetic route was used to

synthesize a diverse set of 2OG derivatives bearing aliphatic

Table 1 Inhibition of AspH by 2OG derivatives

2OG derivative^a

IC₅₀[mM]

2OG derivative^a

IC₅₀[mM]

2OG derivative^a

IC₅₀[mM]

b,c

1

2

Inactive

12

13

0.61 ꢂ 0.09

0.47 ꢂ 0.08

23

24

12.9 ꢂ 1.3

1.2 ꢂ 0.5

Inactive

3

5.7 ꢂ 1.1

48.2 ꢂ 13.1

6.8 ꢂ 0.9

1.6 ꢂ 0.3

2.6 ꢂ 0.8

6.3 ꢂ 2.6

3.6 ꢂ 1.4

4.7 ꢂ 0.2

14

15

16

17

18

19

20^d

21^e

0.51 ꢂ 0.12

0.70 ꢂ 0.11

0.25 ꢂ 0.05

0.43 ꢂ 0.05

0.17 ꢂ 0.03

0.3 ꢂ 0.1

25

26

27

28

29

30

31

32

Inactive

4

5

6

7

8

9

5.2 ꢂ 1.7

10

19.3 ꢂ 1.6

11

Inactive

22

Inactive

33

3.3 ꢂ 1.0

a

All chiral 2OG derivatives were prepared as racemic mixtures. ^bMean of three independent runs (n ¼ 3; mean ꢂ SD). AspH inhibition assays were

c

performed as described in the ESI using 50 nM His₆-AspH_315–758and 1.0 mM hFX-CP_101–119(ESI Fig. S1d) as a substrate. 2OG derivatives were

termed inactive when the IC₅₀-values were >50 mM. The AspH inhibition assays were of good quality which high S/N ratios and Z⁰-factors⁴⁰(>0.5

d

e

for each plate) indicate (ESI Fig. S3). Mixture of racemic diastereomers, dr (cis : trans) ¼ 2.5 : 1. (ꢂ)-(2-Exo,3-endo)-diastereomer.

1330 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

was replaced by rings including heteroaromatic rings (Table 1, (2.5 : 1 mixture of racemic cis/trans-diastereomers) inhibits

entries 22 and 23), aromatic rings (Table 1, entries 24–31), and (IC₅₀ꢃ 5.2 mM; Table 1, entry 20), albeit less eﬃciently than 34.

aliphatic bicyclic rings (Table 1, entry 32). A cyclopropane-con- This might reect the reduced rotational exibility of 35 around

taining 2-oxoacid that was not based on the glutarate C5 skel- the C3–C4 bond due to the presence of the cyclopropane ring.

eton but on the succinate C4 skeleton (48; Table 1, entry 33), was 2OG derivative 36, which is also C3/C4-disubstituted, but which

synthesized.

is signicantly more bulky than 35, inhibits AspH (IC₅₀ꢃ 19.3

The synthesis of 2OG derivatives bearing acid-labile or some mM), but substantially less eﬃciently than does 35 (Table 1,

oxidation-prone moieties was challenging. For example, during entry 21). The thiophene-based 2OG derivative 37 did not

the cyanosulfur ylid oxidation reaction, both ketal and silyl inhibit AspH (Table 1, entry 22). By contrast, the regioisomeric

ether alcohol protecting groups were cleaved and nitrogen- thiophene-based 2OG derivative 38 inhibited AspH with

containing heteroaromatic rings (e.g. pyridines) formed N- moderate eﬃciency (IC₅₀ꢃ 12.9 mM; Table 1, entry 23). This

oxides. Nonetheless, the oxidation conditions were suﬃciently observation might reect the ability of 38, but likely not of 37 to

mild to tolerate alkenes (36; Table 1, entry 21) and substituted better chelate Fe(^II), including that at the AspH active site.

thiophenes (37 and 38; Table 1, entries 22 and 23), which Several Fe(^II)-chelators have been previously identied to inhibit

constitutes an advantage compared to many prior syntheses of AspH in inhibitor screens.^41a

2OG derivatives (Fig. 2).

The phenyl ring regioisomers 39, 40, and 41, which are 2OG

derivatives bearing an aromatic core, did not inhibit AspH

(Table 1, entries 24–26), as it was the case for the other tested

phenyl ring containing 2OG derivatives 42–46 (Table 1, entries

AspH inhibition studies

We then evaluated the potential of the synthesized racemic 2OG 27–31). The 2OG derivative 48 whose carbon scaﬀold was not

derivatives to inhibit AspH by measuring AspH substrate based on glutarate, but derived from succinate, inhibited AspH

depletion and product formation (i.e. by monitoring a +16 Da with moderate eﬃciency (IC₅₀ꢃ 3.3 mM; Table 1, entry 33).

mass shi) using an established solid phase extraction coupled

Interestingly, 4-benzyl-2OG (32) inhibits AspH signicantly

to mass spectrometry (SPE-MS) AspH inhibition assay.⁴¹Half more eﬃciently (IC₅₀ꢃ 0.4 mM; Table 1, entry 17) than its N-

maximum inhibitory concentrations (IC₅₀-values) for all the oxalyl analogue N-oxalyl-^D-phenylalanine (NOFD, IC₅₀ꢃ 15.5

synthetic 2OG derivatives prepared were determined (Table 1). mM),^41awhich is a reported inhibitor of human FIH.²⁴An

3-Methyl-2OG (16) did not inhibit AspH (i.e. IC₅₀> 50 mM), opposite trend was observed for FIH, for which NOFD was

while 2OG derivatives bearing longer carbon chains at the C3- a substantially more eﬃcient inhibitor than 32,²⁴revealing the

position, including 3-ethyl-2OG (17), were eﬃcient inhibitors, context dependent eﬀect of the same 2OG substitutions. The

under the assay conditions. The inhibitory potency decreased structures of these two inhibitors are very similar: the C3

on increasing the length or steric bulk of the carbon chain of the methylene-unit of 4-benzyl-2OG (32) is substituted for an NH-

C3-substituent beyond that of an ethyl group, i.e. C3–Et (17): group in NOFD;²⁴however, 32 was prepared as a racemic

IC₅₀ꢃ 1.2 mM vs. C3–Pr (18): IC₅₀ꢃ 5.7 mM, C3–CH₂Ph (21): IC₅₀mixture whereas NOFD was used in enantiopure ^D-form. To

ꢃ 1.6 mM vs. C3–CH₂CH₂CH₂Ph (20): IC₅₀ꢃ 6.8 mM, and C3– investigate the eﬀect which the NH-group present in NOG and

CH₂CH₂CH₂Ph (20): IC₅₀ꢃ 6.8 mM vs. C3–CH₂CH₂C(CH₃)₃(19): NOFD imposes on AspH inhibition, while excluding possible

IC₅₀ꢃ 48.2 mM. For the C3 benzyl substituted 2OG derivatives interference from the stereochemistry of the inhibitors

21–25 with diﬀerently substituted phenyl rings, the substitution (including with respect of C3-racemisation of the chiral 2-

pattern on the phenyl rings appears to not aﬀect the inhibitor oxoacids), a derivative of 4,4-dimethyl-2OG (34) was thus

potency within experimental error (Table 1, entries 6–10). With synthesized in which the C3 methylene-unit was replaced with

the exception of 4-methyl-2OG (26), which did not inhibit AspH an NH-group (N-oxalyl-a-methylalanine, 49; ESI Fig. S4†). 34

(Table 1, entry 11), 2OG derivatives bearing substituents at the inhibits AspH approximately ten times more eﬃciently (IC₅₀

ꢃ

2OG C4-position (Table 1, entries 12–18) were substantially 0.3 mM; Table 1, entry 19) than 49 (IC₅₀ꢃ 2.9 mM; ESI Fig. S4†).

more potent in inhibiting AspH than those bearing substituents This observation could reect the higher conformational exi-

at the C3-position: their relative potencies increased by a factor bility of 34 and/or the higher stability of the AspH:34 complex.

of 2 (for C3/4-Et) to ꢃ70 (for C3/4-CH₂CH₂C(CH₃)₃). The IC₅₀-

values of the C4-substituted 2OG derivatives range between 0.2

and 0.7 mM (Table 1, entries 12–18) and did not appear to

depend on length or bulk of the tested C4-substituents. The C4-

2OG derivatives compete with 2OG for binding the AspH

active site

substituted 2OG derivatives inhibit AspH with comparable To dene whether the mechanism by which the 2OG derivatives

eﬃciency as the broad-spectrum 2OG oxygenase inhibitor NOG inhibit AspH involves competition with 2OG for binding to the

(3; IC₅₀ꢃ 1.0 mM; ESI Fig. S4†),^41abut less eﬃciently than AspH active site, the eﬀect of altered 2OG concentrations on the

pyridine-2,4-dicarboxylate (2,4-PDCA; IC₅₀ꢃ 0.03 mM),^41awhich IC₅₀-values of AspH was investigated. The IC₅₀-values of four

is another broad-spectrum 2OG oxygenase inhibitor.

potent AspH inhibitors (i.e. 17, 29, 33, and 34) were determined

4,4-Dimethyl-2OG (34) was an eﬃcient AspH inhibitor (IC₅₀at 2OG assay concentrations of 3, 200, 400, and 600 mM (Fig. 3a;

ꢃ 0.3 mM; Table 1, entry 19), whilst the more bulky bicyclo[2.2.2] ESI Table S1†). The results reveal an ascending linear depen-

octane-bearing 2OG derivative 47 did not inhibit (Table 1, entry dence of the IC₅₀-values on the 2OG assay concentration sug-

32). The dimethylcyclopropane-bearing 2OG derivative 35 gesting that the 2OG derivatives inhibit AspH by competing

Chem. Sci., 2021, 12, 1327–1342 | 1331

View Article Online

Chemical Science

Edge Article

Fig. 3 Inhibition of AspH by 2OG derivatives. (a) The AspH IC₅₀-values for the 2OG derivatives 17 (black diamonds), 29 (red squares), 33 (orange

circles), and 34 (green triangles) depend on the 2OG concentration. AspH inhibition assays were performed in triplicate as described in the ESI†

using 3, 200, 400, and 600 mM 2OG. IC₅₀-values are summarized in ESI Table S1;† (b) representative dose–response curves used to determine

IC₅₀-values for the 2OG derivatives 17 (black diamonds), 29 (red squares), 33 (orange circles), and 34 (green triangles) at a 2OG assay

concentration of 3 mM 2OG. Three dose–response curves each composed of technical duplicates were independently determined using SPE-MS

AspH inhibition assays, performed as described in the ESI† and manifest high Z⁰-factors⁴⁰and signal-to-noise ratios (ESI Fig. S3†).

with 2OG for binding to the active site. This is in agreement (ꢃ8%) aer 15 minutes. These results reveal that 3-methyl-2OG

with a Hill coeﬃcient⁴²analysis of the AspH inhibition curves (16) is a substantially more eﬃcient alternative AspH cosub-

which indicates that the 2OG derivatives did not inhibit AspH strate than its isomer 4-methyl-2OG (26), an observation in

by forming colloidal aggregates; the Hill coeﬃcients are in the accord with the observation that C3-substituted 2OG derivatives

range of the expected ‘ideal’ value ꢀ1 (Fig. 3b).⁴³

are generally less eﬃcient AspH inhibitors than their corre-

Neither the position nor the size of the C3/C4-substituent of sponding C4-substituted isomers (Table 1).

the 2OG derivatives had a detrimental eﬀect on the linear

Increasing the steric bulk of the 2OG derivative, while

dependence of their AspH IC₅₀-values on the 2OG concentration maintaining the relative arrangement of the two carboxylate

(Fig. 3a). Eﬃcient inhibition of AspH at higher 2OG assay groups, decreases the catalytic eﬃciency of the cosubstrate

concentrations was observed for 2OG derivatives 33 and 34 (IC₅₀analogue, as revealed by the comparison of the phenyl-ring 2OG

ꢃ 5.6 and 18.7 mM at 0.6 mM 2OG assay concentration, derivative 41 (ꢃ80%) with its bridged bicyclo[2.2.2]octane

respectively; ESI Table S1,† entries 3 and 4).

analogue 47 (ꢃ8%). Derivatives of 41 bearing substituents ortho

to the ketone (e.g. as in 42, ꢃ45%) seemed to be more eﬃcient

AspH cosubstrates than those bearing substituents meta to the

ketone (e.g. as in 43, ꢃ8%). Increasing the size of the substit-

uents ortho to the ketone of 2OG derivative 41 (i.e. ortho-F, 42;

ortho-Br, 44; ortho methyl 46) results in a noticeable decrease in

the eﬃciency to replace 2OG as an AspH cosubstrate (i.e. 42:

ꢃ45%; 44: ꢃ20%; 46: ꢃ5%); all derivatives of 41 were signi-

cantly less eﬃcient with respect to the parent compound

(ꢃ80%).

Some 2OG derivatives can replace 2OG as an AspH cosubstrate

During the assessment of the AspH inhibition data, we observed

that in the presence of two 2OG derivatives, i.e. 3-methyl-2OG

(16) and 4-carboxyphenylglyoxylic acid (41), the extent of

AspH-catalyzed substrate hydroxylation appeared to increase.

We proposed that these 2OG derivatives could replace 2OG and

function as alternative cosubstrates for AspH. AspH substrate

hydroxylation was thus investigated in the absence of 2OG: high

levels of AspH substrate hydroxylation were observed at elevated

concentrations of the 2OG derivatives 16 (>95%) and, somewhat

unexpectedly, the phenyl ring derivative 41 (ꢃ80%) in the

absence of 2OG, demonstrating that these two 2OG derivatives

can replace 2OG as an AspH cosubstrate (Fig. 4a–c). NMR

studies revealed that AspH converts the 2OG derivatives 16 and

41 into 2-methylsuccinate (50) and terephthalate (51), respec-

tively, in an analogous manner to which it converts 2OG into

succinate (Fig. 4a–c; ESI Fig. S5†).

Kinetic analyses of the two most eﬃcient 2OG substitute

AspH cosubstrates identied (i.e. 16 and 41) were performed.

Maximum velocities (v^a_m^p_a^p_x) and Michaelis constants (K^a_m^pp) were

determined employing SPE-MS turnover assays, albeit under

modied conditions than previously reported for 2OG as

a cosubstrate (Fig. 4d–f),²³as L-ascorbic acid (LAA), which is

commonly added to 2OG oxygenase assays, was included in the

assay buﬀer. The presence of LAA aﬀected the kinetic parame-

ters for 2OG when compared to the previous parameters (Table

app

max

2, entry 1). The v

(2OG) did not change signicantly, being

ꢃ16.8 ꢄ 10^ꢀ3mM s^ꢀ1in the absence of LAA²³and ꢃ15.0 ꢄ 10^ꢀ3

mM s^ꢀ1in the presence of LAA. The K^a_m^pp(2OG)-value in the

presence of LAA was approximately double (ꢃ1.3 mM) compared

to that in the absence of LAA (ꢃ0.6 mM).²³However, this

K^a_m^pp(2OG)-value is still in the range of those values reported for

most other human 2OG oxygenases including for the HIF-

a prolyl hydroxylases and FIH (1–25 mM)⁴⁴and bovine AspH (ꢃ5

mM).⁴⁵

To investigate whether the other synthesized 2OG derivatives

can substitute for 2OG as a cosubstrate for AspH-catalyzed

hydroxylation, assays were performed in the absence of 2OG,

but in the presence of high concentrations (330 mM) of all the

synthetic 2OG derivatives (ESI Table S2†). In addition to the

2OG derivatives 16 and 41, AspH substrate hydroxylation was

observed for the 2OG derivatives 26 (ꢃ10%), 38 (ꢃ15%), 42

(ꢃ45%), 43 (ꢃ8%), 44 (ꢃ20%), 45 (ꢃ8%), 46 (ꢃ5%), and 47

1332 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

Fig. 4 AspH steady-state kinetic parameters for selected 2OG derivatives measuring initial hydroxylation rates of a synthetic thioether linked

cyclic peptide AspH substrate. (a) AspH catalyzes the oxidative decarboxylation of 3-methyl-2OG (16) to give 2-methylsuccinate (50); (b) AspH

catalyzes the oxidative decarboxylation of 4-carboxyphenylglyoxylic acid (41) to give terephthalate (51); (c) the 2OG derivatives 16 and 41 replace

2OG as an AspH cosubstrate, as monitoring AspH substrate hydroxylation manifests. AspH assays were performed as described in the ESI† using

0.1 mM AspH, 2 mM hFX-CP_101–119(ESI Fig. S1d†), 50 mM FAS, 330 mM 2OG or 2OG derivative in 50 mM HEPES (pH 7.5, 20 ^ꢁC); (d) K_m^appof AspH for

2OG; (e) K^a_m^ppof AspH for 16; (f) K^a_m^ppof AspH for 41. AspH assays were performed as described in the ESI,† data are shown as the mean of three

independent runs (n ¼ 3; mean ꢂ standard deviation, SD). The results are summarized in Table 2 and the peptide hydroxylation rates are shown in

ESI Fig. S6.†

Compared to the K_m^app-value of AspH for 2OG (ꢃ1.3 mM; Table aﬃnity of AspH for 16 compared to 2OG. By contrast, the K^a_m^pp

-

2, entry 1), the K^a_m^pp-value for 3-methyl-2OG (16) was about ve value of AspH for the 2OG derivative 41 was ꢃ47 times higher

times lower (ꢃ0.3 mM; Table 2, entry 2), indicating a higher (ꢃ62 mM; Table 2, entry 3) than that for 2OG, indicating much

Table 2 Steady-state kinetic parameters for AspH. Maximum velocities (v ), Michaelis constants (K^a_m^pp), turnover numbers (k_cat), and speciﬁcity

app

max

constants (k_cat/K_m) of His₆-AspH_315–758for 2OG and the 2OG derivatives 16 and 41^a,b

app

max

AspH co-substrate

v

[mM s^ꢀ1

]

K^a_m^pp[mM]

k_cat[s^ꢀ1

]

k_cat/K_m[mM^ꢀ1s^ꢀ1

]

1

2

3

2OG

15.0 ꢄ 10^ꢀ3ꢂ 0.9 ꢄ 10^ꢀ3

6.9 ꢄ 10^ꢀ3ꢂ 0.2 ꢄ 10^ꢀ3

4.0 ꢄ 10^ꢀ3ꢂ 0.3 ꢄ 10^ꢀ3

1.3 ꢂ 0.3

0.27 ꢂ 0.04

61.9 ꢂ 11.0

0.17 ꢂ 0.03

0.08 ꢂ 0.01

0.04 ꢂ 0.01

0.13 ꢂ 0.03

0.30 ꢂ 0.06

3-Methyl-2OG (16)

4-Carboxyphenylglyoxylic acid (41)

0.65 ꢄ 10^ꢀ3ꢂ 0.2 ꢄ 10^ꢀ3

a

Mean of three independent runs (n ¼ 3; mean ꢂ SD). ^bAspH assays were performed as described in the ESI using 0.1 mM His₆-AspH_315–758and 2.0

mM hFX-CP_101–119(ESI Fig. S1d) as a substrate.

Chem. Sci., 2021, 12, 1327–1342 | 1333

View Article Online

Chemical Science

Edge Article

less eﬃcient binding. All three K^a_m^pp-values range signicantly residues rather than by the typical triad of ligands (HXD/E/H)

below reported 2OG concentrations in healthy cells (up to >1 found in other human 2OG hydroxylases.^7,14Superimposition of

mM),⁴⁶which however vary substantially, but are in the the AspH:2OG structure with the reported AspH:NOG²²and

approximate range of reported physiological 2OG levels in AspH:^L-malate²²structures reveals that AspH adopts similar

human plasma (9–12 mM 2OG).⁴⁷

conformations in all structures and that 2OG binds the AspH

Based on the determined concentration of active AspH (90.8 active site in a similar manner to NOG (Ca RMSD ¼ 0.21 and

˚

ꢂ 13.7 nM for an original estimated AspH assay concentration 0.21 A, ESI Fig. S7†).

of 100 nM AspH),²³turnover numbers (catalytic constants, k_cat

)

To investigate the eﬀects of substrate binding in the pres-

and specicity constants (k_cat/K_m) were calculated for the 2OG ence of 2OG, AspH was crystallized in the presence of 2OG,

derivatives (Table 2). Comparison of the k_cat-values reveals that Mn(^II), and the synthetic hFX-EGFD1_86–124-4Ser substrate²²(ESI

the impact of the 2OG derivatives on k_cat-values is notably Fig. S1c†), which mimics the EGFD1 of the reported AspH

smaller than on K^a_m^pp-values which indicates that eﬃcient AspH- substrate human coagulation factor X (hFX).⁴⁹The structure was

catalyzed substrate hydroxylation is still feasible with the 2OG solved by molecular replacement using a reported structure

derivatives (Table 2): for AspH, the k_cat-value for 16 (ꢃ0.08 s^ꢀ1

;

(PDB ID: 5JTC)²²as a search model (P2₁2₁2₁space group; 2.3 A

˚

Table 2, entry 2) was about half the k_catfor 2OG (ꢃ0.17 s^ꢀ1

resolution). The active site region manifested electron density

Table 2, entry 1), whereas the k_catfor 41 (ꢃ0.04 s^ꢀ1; Table 2, for both 2OG (Fig. 5b) and for the hFX-EGFD1_86–124-4Ser peptide

entry 3) was about a quarter of that for 2OG. Comparison of the (ESI Fig. S8†). Substrate binding to AspH aﬀects the relative

k_cat/K_m-values clearly reveals the potential of 2OG derivatives to alignment of the oxygenase and TPR domains, i.e. the distance

substitute for 2OG itself, with racemic 16 being of similar eﬃ- between the Ca atoms of Leu433 on TPR repeat a6 and Pro756

˚

ciency to 2OG. Although 41 is a much less eﬃcient substrate, its in the AspH C terminal region decreases from ꢃ20 A to ꢃ14 A

conversion reveals the potential for unexpected cosubstrate upon substrate binding (ESI Fig. S9†). Evidence for an induced

utilization by 2OG oxygenases.

t substrate binding mechanism involving this conformational

change has been described when the cosubstrate 2OG was

substituted for NOG.²²The signicant conformational changes

in the AspH oxygenase domain triggered by substrate binding

Crystallography

The AspH turnover assays indicated that the synthetic 2OG (ESI Fig. S10†) do not aﬀect the observed mode of 2OG binding

derivatives compete with 2OG for binding to the AspH active in the active site (Fig. 5c); the binding modes of both 2OG and

site. To investigate the divergent eﬀects of C3/C4-substituted NOG are very similar, both in the presence or absence of

2OG derivatives on AspH catalysis, i.e. AspH inhibition or substrate (ESI Fig. S9†). Interestingly, whilst in the 2OG complex

promoting AspH activity, crystallographic studies were initi- structure the substrate residue Asp103_hFXwas observed in

ated. In the reported AspH crystal structures, the natural AspH a single conformation whereas in the analogous NOG structure

cosubstrate 2OG was substituted for a 2OG competing inhibitor it was observed in two conformations (ESI Fig. S9†).

(e.g. NOG, 2,4-PDCA or ^L-malate),^22,41abut an AspH crystal

structure complexed with 2OG has not previously been methyl-2OG (16) and Mn(^II), both with and without the hFX-

reported.

EGFD1_86–124-4Ser substrate bound, were obtained

High resolution crystal structures of AspH complexed with 3-

To enable comparisons of how 2OG and the 2OG derivatives (AspH:16:hFX-EGFD1_86–124-4Ser, ESI Fig. S11;† AspH:16, ESI

˚

bind AspH, AspH was rst crystallized in the presence of 2OG Fig. S12;† 1.5 and 1.8 A resolution, respectively). Substitution of

with the natural AspH cofactor Fe(^II) being replaced by Mn(II). 2OG for 16 did not trigger signicant changes in the AspH

AspH crystallized in the absence of substrate in the P2₁2₁2₁conformations (superimposition of the AspH:2OG and the

˚

space group (AspH:2OG; 2.1 A resolution), the structure, as were AspH:16 structures: Ca RMSD ¼ 0.38 A; superimposition of the

the subsequently described structures, was solved by molecular AspH:2OG:hFX-EGFD1_86–124-4Ser and the AspH:16:hFX-

replacement using a reported AspH structure (PDB ID: 5JZA)²²as EGFD1_86–124-4Ser structures: Ca RMSD ¼ 0.23 A; ESI Fig. S12

˚

a search model (ESI Fig. S7†). Clear electron density corre- and S13†). A similar change in the relative alignment of the

sponding to 2OG was observed (Fig. 5a); the C5-carboxylate of AspH oxygenase and TPR domains upon substrate binding was

2OG being positioned to form a salt bridge with the side chain observed when 2OG was substituted for 16 (ESI Fig. S13†).

˚

of Arg735 (2.4 and 3.0 A) and to interact with Ser668 (2.6 A),

Although 3-methyl-2OG (16) was prepared and used as

which is part of an ‘RXS motif’ present in some other 2OG a racemic mixture, the electron density map corresponded to

oxygenases.⁴⁸The C1-carboxylate of 2OG is positioned to the, at least predominant, presence of the (R)-enantiomers at

˚

interact with Arg688 (2.7 and 2.9 A) and His690 (3.3 A). The C1- the active site in both the AspH:16 and AspH:16:hFX-EGFD1_86–

carboxylate of 2OG and the 2OG C2-carbonyl group complex the ₁₂₄-4Ser structures (Fig. 5d and e; ESI Fig. S11 and S12†). The

˚

Mn ion in a bidentate manner (1.6 and 2.5 A; Fig. 5a). Two water carboxylate groups of the (R)-enantiomer of 16 interact with the

˚

molecules also coordinate the Mn ion (2.1 and 2.4 A) along with same AspH residues as 2OG in both the AspH:16 and

˚

the two anticipated residues His679 and His725 (2.2 and 2.1 A; AspH:16:hFX-EGFD1_86–124-4Ser structures. It is positioned to

˚

Fig. 5a). Thus, the AspH:2OG structure supports the proposal salt bridge with Arg735 (2.6/2.8 and 2.7/2.7 A, respectively) with

(based on the AspH structures in complex with the 2OG its C5-carboxylate and is positioned to interact with Ser668 (2.5

analogue NOG²²) that the active site metal, when bound to the and 2.6 A, respectively) through its C5-carboxylate, with His690

˚

natural AspH cosubstrate 2OG, is complexed by only two AspH (2.8 and 2.8 A, respectively) and Arg688 (2.8 and 2.9 A,

1334 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

Fig. 5 The conformations of 3-methyl-2OG (16) and 2OG diﬀer at the AspH active site. Colors: grey: His₆-AspH_315–758; green: carbon-backbone

of 2OG; yellow: carbon-backbone of (R)-16; violet: Mn; orange: carbon-backbone of the hFX-EGFD1_86–124-4Ser peptide (ESI Fig. S1c†); red:

oxygen; blue: nitrogen; w: water. (a) Representative OMIT electron density map (mF_o-DF_c) contoured to 5s around 2OG of the AspH:2OG

structure; (b) representative OMIT electron density map (mF_o-DF_c) contoured to 5s around 2OG of the AspH:2OG:hFX-EGFD1_86–124-4Ser

structure. The 2OG C5-carboxylate is positioned to salt bridge with the Arg735 side chain (2.5 and 2.9 A˚) and to interact with Ser668 (2.5 A˚). 2OG

is positioned to interact with His690 (2.9 A˚) and Arg688 (2.8 A˚) through its C1-carboxylate, and with the Mn ion via its C1-carboxylate (2.2 A˚) and

C2-carbonyl (2.3 A˚). The Mn ion is also complexed by His679 (2.2 A˚), His725 (2.1 A˚), a water molecule (2.1 A˚), and the C4-carboxylate of the

Asp103_hFXpeptide substrate (2.4 A˚); (c) superimposition of views from the AspH:2OG:hFX-EGFD1_86–124-4Ser structure with one from the

AspH:2OG structure (beige: His₆-AspH_315–758; lavender: Mn ion; lemon: carbon-backbone of 2OG); (d) representative OMIT electron density

map (mF_o-DF_c) contoured to 5s around (R)-16 of the AspH:16 structure; (e) representative OMIT electron density map (mFo-DFc) contoured to

5s around (R)-16 of the AspH:16:hFX-EGFD1_86–124-4Ser structure; (f) superimposition of views from the AspH:16:hFX-EGFD1_86–124-4Ser

structure with one from the AspH:2OG:hFX-EGFD1_86–124-4Ser structure (pale green: His₆-AspH_315–758; green: carbon-backbone of 2OG;

lavender: Mn ion; aquamarine: carbon-backbone of hFX-EGFD1_86–124-4Ser).

respectively) through its C1-carboxylate, and with Mn(II) Fig. 5f), which together with Met670 form one face of a hydro-

˚

through its C1-carboxylate (2.2 and 2.1 A, respectively) and C2- phobic pocket, to which the indole ring of Trp625 also

˚

carbonyl groups (2.2 and 2.1 A, respectively). Notably, super- contributes (ESI Fig. S16c†).

imposition of the AspH:16:hFX-EGFD1_86–124-4Ser and

The crystallographically observed diﬀerent conformations

AspH:2OG:hFX-EGFD1_86–124-4Ser structures (ESI Fig. S12†) which (R)-16 and 2OG occupy when bound to AspH may reect

reveals that the (R)-enantiomer of 16 adopts a diﬀerent the diﬀerences in their kinetic parameters. Thus, the hydro-

conformation than 2OG at the active site. Whilst the oxalyl- phobic interactions that the C3-methyl of (R)-16 forms with the

groups of both 2OG and (R)-16 bind the metal in an identical side-chains of Val676 and Val727 might increase its binding

manner and the two C5-carboxylate groups are superimposable, aﬃnity, in agreement with its K^a_m^pp-value that is about ve time

the C3- and C4-methylenes adopt diﬀerent conformations lower than that for 2OG (Table 2). The k_cat-value for (racemic) 16

(Fig. 5f). The C3-methyl group of (R)-16 faces towards AspH is about half the k_cat-value for 2OG (Table 2, entries 1 and 2),

residues Val676 and Val727 (distance of the methyl C-atom of possibly reecting the crystallographic evidence that the (R)- but

˚

(R)-16 to the g-methyl C-atoms of the Val residues: ꢃ3.5–4.6 A; not the (S)-enantiomer of 16 is a cosubstrate for AspH. It should

Chem. Sci., 2021, 12, 1327–1342 | 1335

View Article Online

Chemical Science

Edge Article

also be noted that the (S)- and (R)-enantiomers of 16 will Fig. S14†), even though 17 was used as a racemic mixture during

interconvert in aqueous media and in the presence of metal crystallization. The (R)-enantiomers of 17 and 3-methyl-2OG

ions (ESI† Sections 4 and 5)⁵⁰and that, as (R)-16 is consumed in (16) occupy similar conformations when bound to the AspH

assays, this interconversion may become rate-limiting.

active site (Fig. 6c), with their C3-substituents facing towards

Next, AspH was crystallized in the presence of the inhibitor 3- the side chains of Val676 and Val727. Superimposition of the

ethyl-2OG (17); a crystal structure of AspH complexed with 17, AspH:17:hFX-EGFD1_86–124-4Ser, the AspH:2OG:hFX-EGFD1_86–

Mn(^II), and the hFX-EGFD1_86–124-4Ser substrate was obtained to ₁₂₄-4Ser, and the AspH:16:hFX-EGFD1_86–124-4Ser structures

˚

1.8 A resolution (AspH:17:hFX-EGFD1_86–124-4Ser, ESI Fig. S14†). reveals no signicant changes in the AspH and the hFX-

As with 16, the results suggest that 17 competes with 2OG for EGFD1_86–124-4Ser substrate conformations (ESI Fig. S15†), and

binding the AspH active site, in agreement with inhibition the presence of the diﬀerent 2OG analogues does not appear to

assays with varied 2OG concentrations (Fig. 3a and ESI Table aﬀect the conformations of AspH active site side-chain residues

S1†). As for 16, the observed electron density corresponded to (Fig. 6b and c).

the presence of the (R)-enantiomer of 17 (Fig. 6a and ESI

Fig. 6 Binding of 3-ethyl-2OG (17) and 4,4-dimethyl-2OG (34) at the AspH active site. Colors: grey: His₆-AspH_315–758; salmon: carbon-

backbone of (R)-17; slate blue: carbon-backbone of 34; violet: Mn; orange: carbon-backbone of the hFX-EGFD1_86–124-4Ser peptide (ESI

Fig. S1c†); red: oxygen; blue: nitrogen; w: water. (a) Representative OMIT electron density map (mF_o-DF_c) contoured to 3s around (R)-17 of the

AspH:17:hFX-EGFD1_86–124-4Ser structure. (R)-17 binds similarly to (R)-16 (ESI Fig. S14a†); (b) superimposition of views of the AspH:17:hFX-

EGFD1_86–124-4Ser and the AspH:2OG:hFX-EGFD1_86–124-4Ser (pale green: His₆-AspH_315–758; lavender: Mn ion; green: carbon-backbone of 2OG)

structures; (c) superimposition of active sites views of the AspH:17:hFX-EGFD1_86–124-4Ser and the AspH:16:hFX-EGFD1_86–124-4Ser (pale yellow:

His₆-AspH_315–758; lavender: Mn ion; yellow: carbon-backbone of (R)-16) structures; (d) representative OMIT electron density map (mF_o-DF_c)

contoured to 3s around 34 of the AspH:34:hFX-EGFD1_86–124-4Ser structure. 34 binds similarly to 2OG (ESI Fig. S16a†); (e) superimposition of

views of the AspH:34:hFX-EGFD1_86–124-4Ser and the AspH:2OG:hFX-EGFD1_86–124-4Ser (pale green: His₆-AspH_315–758; lavender: Mn ion; green:

carbon-backbone of 2OG) structures; (f) superimposition of active site views of the AspH:34:hFX-EGFD1_86–124-4Ser and the AspH:17:hFX-

EGFD1_86–124-4Ser (brown: His₆-AspH_315–758; lavender: Mn ion; salmon: carbon-backbone of 17) structures.

1336 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

We also crystallized AspH in the presence of 4,4-dimethyl- dehydrogenases^51a,51b,52that employ 2OG as a substrate or

2OG (34), Mn(^II), and the hFX-EGFD1_86–124-4Ser substrate (2.3 cosubstrate, but has been employed to a much lesser extent with

-

2OG oxygenases.^24–26By contrast with 2OG derivatives bearing

˚

A resolution; ESI Fig. S16†). In this AspH:34:hFX-EGFD1_86–124

4Ser structure, clear electron density for 34 was observed in the major structural modications of the 2-oxo-1,5-dicarboxylic

AspH active site (Fig. 6d). The pro-R methyl group of 34 projects acid scaﬀold,⁵³the synthetic accessibility of C3/C4-substituted

towards one face of the hydrophobic pocket (Val626, Val727, 2OG derivatives has been hitherto limited, a factor which

Met670), whereas the pro-S methyl group of 34 is in proximity of might have hampered detailed biochemical studies on their

Met670 and faces towards the indole ring of Trp625 (Fig. 6d and eﬀects on 2OG oxygenases. To address this need, we developed

ESI Fig. S16c†).

an eﬃcient synthesis of C3/C4-substituted 2OG derivatives

While the AspH and the hFX-EGFD1_86–124-4Ser substrate relying on the use of cyanosulfur ylids^33a,34as key intermediates

conformations of the AspH:34:hFX-EGFD1_86–124-4Ser structure (Scheme 1). Our synthesis compares favorably to reported

do not diﬀer from those of previous structures (ESI Fig. S17†), syntheses of 2OG derivatives (Fig. 2);^25,27–30it is scalable, aﬀords

the conformations of 34, of (R)-3-ethyl-2OG (17), and of N-oxalyl- the 2OG derivatives and their synthetic precursors in high

a-methylalanine (49), which is a derivative of 34, diﬀer purity suitable for biochemical applications, and avoids the use

substantially in the crystalline state (Fig. 6f, ESI Fig. S18 and of strong bases, acids, and oxidants. The broad substrate scope

S19†). By contrast, the conformations of 2OG and 34 are very of the synthesis reects the mild reaction conditions; for

similar (Fig. 6e); we therefore investigated by NMR if 34 is example, 2OG derivatives bearing oxidation-prone olens and

converted into 2,2-dimethylsuccinate, in the presence of thiophenes were readily synthesized (36–38; Table 1). No special

substantially higher AspH concentrations compared to the laboratory equipment, such as an ozone generator, is required

original SPE-MS assay conditions (ꢃ100 times more AspH). The for the synthesis rendering it particularly user-friendly. We

results reveal that 34 indeed undergoes slow AspH-catalyzed prepared racemic mixtures of C3/C4-substituted 2OG deriva-

oxidative decarboxylation, whereas another AspH inhibitor, 4- tives, however, the same synthetic strategy could be applied for

benzyl-2OG (32), was apparently not turned over under the same the synthesis of enantiopure C4-substituted 2OG derivatives

reaction conditions (ESI Fig. S20†).

using enantiopure mono-methyl dicarboxylic acid half-esters 13

Thus, the combined crystallographic analyses reveal that the as starting materials, which, for example, can be obtained by

C3/C4-substituted 2OG derivatives bind AspH in the same asymmetric hydrogenation reactions from itaconates.⁵⁴The C3/

general manner as 2OG, with near identical binding modes for C4-substituted 2OG derivatives were obtained as dicarboxylic

the oxalyl-groups and the C5-carboxylates. The diﬀerent acids (6, Scheme 1) and dimethyl dicarboxylates (12, Scheme 1).

conformations for the C3- and C4-methylenes observed, The former are useful for experiments with isolated enzymes as

however, do not correlate with catalysis versus inhibition performed in this study, whereas the latter can be used for cell-

(assuming the crystallographic binding modes reect those in based and in vivo experiments, due to their improved cell-wall

solution). Thus, whilst the C3- and C4-methylenes of (R)-16 and penetrating abilities.⁵⁵

(R)-17 adopt a very similar conformation, which is distinct from

that of 2OG and 34, 16 is a cosubstrate, whereas 17 is an inhibited human AspH by a mechanism involving competition

inhibitor. with 2OG for binding the active site as revealed by inhibition

Many of the C3/C4-substituted 2OG derivatives synthesized

Hydrophobic interactions made by the C3-methyl and C3- assays performed at variable 2OG concentrations and crystal-

ethyl groups of (R)-16 and (R)-17, respectively, and of one of lographic studies (Fig. 3 and 6). In general, the C4-substituted

the methyl groups of 34 with AspH, are consistent with the 2OG derivatives were more eﬃcient AspH inhibitors than the

tighter binding of 16 as compared with 2OG and judged by C3-substituted 2OG derivatives (Table 1). The C4-substituted

K^a_m^ppcomparison (Table 2). However, the k_catfor 16 is approxi- 2OG derivatives were also more potent AspH inhibitors than

mately half that of 2OG whereas 17 and 34 are inhibitors and/or the corresponding C4-substituted NOG derivatives (ESI

poor substrates, respectively (Table 2); the reason for these Fig. S4†). NOG and other N-oxalyl amino acids are plant natural

diﬀerences is uncertain, but it may reect slower binding of O₂, products and it is proposed that they may act as enzyme

or a slower subsequent step during catalysis, e.g. reaction of the inhibitors in vivo.⁵⁶Our observations thus raise the possibility

ferryl intermediate proposed to be present in the 2OG oxygenase that naturally occurring 2-oxoacids may be biologically relevant

catalytic cycle or release of the succinate coproduct (ESI modulators of the activities of 2OG oxygenases and related

Fig. S21†). Given the reduced k_catfor 16, it is reasonable to enzymes.

propose that the ‘k_cat’ for the C3-ethyl substituted inhibitor 17

Our results indicate that some 2OG derivatives eﬃciently

will be reduced even further, potentially approaching zero, inhibit AspH at physiologically relevant 2OG levels which range

potentially as a consequence of particular strong interaction from 9–12 mM 2OG in human plasma⁴⁷to >1 mM 2OG in cells.⁴⁶

with the hydrophobic pocket.

For example, 2OG derivative 33 inhibits AspH with 0.6 mM 2OG

in the assay (IC₅₀ꢃ 5.6 mM; ESI Table S1†). 2OG competitive

AspH inhibitors, such as 33, 34, or optimized variants of them,

might be useful from a therapeutic perspective, because their

Discussion

The use of C3/C4-substituted 2OG derivatives has been inhibitory eﬀect is unlikely to be compromised by mutations.

a productive strategy to investigate inter alia the function, Indeed an AspH mutation associated with Traboulsi syndrome

mechanism, and inhibition of aminotransferases^30,51and

Chem. Sci., 2021, 12, 1327–1342 | 1337

View Article Online

Chemical Science

Edge Article

occurs in the 2OG binding site (i.e. R735W) and is likely possible that 41 or related aromatic compounds may modulate

inactivating.^20a

the catalytic activity of 2OG oxygenases in vivo.

At high enzyme concentrations, AspH converts some of the

The k_cat/K_m-value of AspH for 3-methyl-2OG (16) is approxi-

inhibitors (e.g. 34) slowly into the corresponding succinate mately three times higher than for 2OG, suggesting the feasi-

derivatives (ESI Fig. S20†), indicating that the hydrophobic bility of 2OG derivatives to selectively enhance AspH (or indeed

interactions of the 2OG C3/C4-substituents with AspH stabilize other 2OG oxygenase) catalysis in cells or on the cell surface of

the AspH:2OG derivative complexes and/or reduce the rota- cancer cells in the presence of 2OG (Table 2). 16 is of particular

tional exibility of the cosubstrate necessary to enable its interest because it is a reported ingredient of human nutrition

oxidative decarboxylation. This observation is interesting (i.e. it is present in honey);⁶⁷thus, 16 might modulate the

because in some contexts, e.g. the inhibition of HIF-a prolyl activity of AspH and potentially other 2OG oxygenases in

hydroxylases, compounds that do not completely block activity humans. 2-Methylsuccinate, which is formed by the AspH-

even when present in excess, including poor cosubstrates, may catalyzed oxidative decarboxylation from 16 (ESI Fig. S5†), has

actually be desirable, as they may help avoid overdose.⁵⁷

been detected in human urine⁶⁸and is inter alia used as

C3/C4-substituted 2OG derivatives bear the potential to be a biomarker for metabolic diseases as it is a product of other

used as small-molecule probes^55ain cells or in vivo to modulate metabolic pathways (i.e. isoleucine catabolism);⁶⁹this, however,

the catalytic activity of 2OG oxygenases, provided selective does not rule out the possibility that some 2-methylsuccinate

interaction can be achieved. In this regard, the 2OG derivatives might originate from the oxidative decarboxylation of 16 cata-

might, at least partially, be selective AspH inhibitors consid- lyzed by 2OG oxygenases. 16 is also a proposed precursor of 3-

ering that C4-substituted 2OG derivatives were reported to not methyl glutamate, which is incorporated in natural products

inhibit wild-type KDM4A,²⁵which is a 2OG dependent histone such as polytheonamide A and B⁷⁰and daptomycin.⁷¹16 is likely

demethylase.⁵⁸Note that, NOG derivatives have been used in biosynthesized in microorganisms through the direct reaction

vitro and in cellular experiments to modulate the catalytic of 2OG and SAM,⁷²suggesting that it might also be bio-

activities of FIH²⁴as well as of the KDM4 ⁵⁹and ten-eleven- synthesized by animals in a similar manner. The results thus

translocation (TET) enzymes⁶⁰with some selectivity. Our raise the possibility that 16 or other C3/C4-substituted 2OG

results indicate that C4-substituted 2OG derivatives are more derivatives, including their corresponding glutamate deriva-

eﬃcient in inhibiting AspH than the corresponding C4- tives, are human/animal metabolites and/or are bioavailable

substituted NOG derivatives (ESI Fig. S4†), as opposed to through nutrition or the gut microbiome.

human FIH which also catalyzes the hydroxylation of Asp- and

In comparison with 16, its isomer 4-methyl-2OG (26), which

Asn-residues and for which NOG derivatives were more potent is a reported metabolite in plants⁷³and ingredient of wine,⁷⁴is

inhibitors.²⁴With regard to enzymes other than 2OG oxy- substantially less eﬃcient in substituting for 2OG as a cosub-

genases, the AspH inhibitor 4,4-dimethyl-2OG (34), for example, strate (ESI Table S2†), highlighting that the position of the 2OG

does not eﬃciently substitute for 2OG in glutamic oxaloacetic substituent determines the ability of the 2OG derivatives to

aminotransferase catalyzed transamination reactions suggest- serve as an alternative AspH cosubstrate.

ing that it might not interfere with the catalytic activities of

human aminotransferases.⁶¹

AspH was co-crystallized in the presence of 3-methyl-2OG

(16) aﬀording structures with the highest resolution reported

˚

The lysine metabolite 2-oxoadipate, which is based on a C6 for AspH so far (1.5 and 1.8 A; ESI Fig. S11 and S12†). The

rather than a C5 carbon skeleton as in 2OG, is capable of acting crystallographic studies revealed that 16 substitutes for, and

as a relatively poor cosubstrate for procollagen prolyl hydroxy- binds similarly to, 2OG in the AspH active site (Fig. 5). However,

lases,⁶²phytanoyl-CoA 2-hydroxylase (PAHX),⁶³and for a bacte- the C3- and C4-methylenes of 16 occupy diﬀerent conforma-

rial ethylene-forming 2OG dependent enzyme.⁶⁴However, to tions orienting the 2OG C3-methyl-substituent towards Val676

our knowledge, C3/C4-substituted 2OG derivatives have so far and Val727, so enabling hydrophobic interactions. Superim-

not been reported to substitute for 2OG as a cosubstrate for position of 16 and 3-ethyl-2OG (17), which is a potent AspH

wild-type 2OG oxygenases. The observation that several of our inhibitor, reveals that both 2OG derivatives adopt a similar

synthetic 2OG derivatives eﬃciently substituted for 2OG as an conformation when bound to AspH (Fig. 6c). Despite the use of

AspH cosubstrate, with 2OG derivatives 16 and 41 being the racemic 16 and 17 for crystallizations, in both cases the (3R)-

most eﬃcient (ESI Table S2†), is therefore of more general enantiomers were observed by crystallography, with the C3-alkyl

interest.

substituents interacting with one face of a hydrophobic pocket

The observation that 4-carboxyphenylglyoxylic acid (41) can (ESI Fig. S16c†). Thus, subtle changes in the structure of the

promote turnover of a 2OG oxygenase is remarkable because of 2OG derivative can have a pronounced eﬀect on AspH catalysis,

its cyclic aromatic scaﬀold. 41 is thus a promising candidate to resulting in either eﬃcient substrate hydroxylation or eﬃcient

modulate AspH activity in vivo, as it might display selectivity for AspH inhibition. The exact factors that determine whether

AspH over other 2OG oxygenases because of its distinctive a particular 2OG derivative inhibits or enables AspH catalysis

structure. The oxidative decarboxylation of 41 by AspH is are unknown, but might relate to diﬀerences in oxygen binding,

reminiscent of the reaction catalyzed by 4-hydroxyphenyl pyru- modulation of steps aer oxygen binding or to the stability of

vate dioxygenase (HPPD),⁶⁵which is from a diﬀerent class of the AspH:succinate derivative complexes (ESI Fig. S21†).

Fe(^II)-dependent oxygenases. A regioisomer of 41, 3-carbox-

The potential of compensating catalytically inactivating

yphenylglyoxylic acid (40), is a plant metabolite;⁶⁶thus, it is (with 2OG as a cosubstrate) mutants of 2OG oxygenases with 2-

1338 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

oxoacids has been exemplied in the case of PAHX.⁶³Our results

thus raise to possibility that naturally occurring 2OG

derivatives/analogues may modulate the activity of 2OG oxy-

genases and related enzymes (e.g. HPPD⁶⁵), either by acting as

inhibitors or by replacing 2OG in catalysis. Indeed, it is possible

that some of the 2OG oxygenases without assigned biochemical

Acknowledgements

We thank the Wellcome Trust (106244/Z/14/Z), Cancer Research

UK (C8717/A18245), and the Biotechnology and Biological

Sciences Research Council (BB/J003018/1 and BB/R000344/1)

for

funding.

L.

B.

thanks

the

Deutsche

For-

functions (e.g. Jumonji

C domain-containing protein 1c,

schungsgemeinscha for a fellowship (BR 5486/2-1). Y. N.

thanks JSPS for an Overseas Research Fellowship (2020060219)

and the Daiichi Sankyo Foundation of Life Science. We thank

the Diamond Light Source and staﬀ for allocation of beam time

and support. We thank the MS and NMR facilities from the

Chemistry Research Laboratory, University of Oxford, for

excellent support.

JMJD1C)⁷⁵or with unusual Fe(II) binding site geometries (e.g.

PHD nger protein 2 (PHF2)⁷⁶and hairless⁷⁷) use 2-oxoacid

cosubstrates other than 2OG. Future work will focus on

exploring these possibilities using the 2OG derivatives

described here and others prepared by the cyanosulfur ylid

methodology.

References

Conclusions

´

1 (a) W. L. Araujo, A. O. Martins, A. R. Fernie and T. Tohge,

A user-friendly eﬃcient synthesis of C3/C4-substituted 2OG

derivatives based on the use of cyanosulfur ylids was developed

and employed to aﬀord a diverse set of 2OG derivatives for

detailed biochemical and structural investigations on the

cosubstrate selectivity of the human 2OG oxygenase AspH. The

overall results reveal that C3/C4-substituted 2OG derivatives can

have profound eﬀects on AspH catalysis and, by implication,

likely other 2OG oxygenases. Simple alkyl substituents, e.g.

methyl at the 2OG C3- or C4-postion, enables retention of

productive catalysis, likely in a stereoselective manner. By

contrast, e.g. 2OG C4-dimethylation leads to inhibition/low

levels of cosubstrate activity. The use of 2OG derivatives may

thus inform on both the catalytic mechanisms and biological

roles of AspH and other 2OG oxygenases, and aid in the devel-

opment of new types of small-molecules that modulate 2OG

oxygenase activity.

˙

´

Front. Plant Sci., 2014, 5, 552; (b) B. Zdzisinska, A. Zurek

and M. Kandefer-Szerszen, Arch. Immunol. Ther. Exp., 2017,

´

65, 21–36.

2 (a) A. J. Ninfa and P. Jiang, Curr. Opin. Microbiol., 2005, 8,

168–173; (b) M. Lancien, P. Gadal and M. Hodges, Plant

Physiol., 2000, 123, 817–824.

3 (a) L. Dang, D. W. White, S. Gross, B. D. Bennett,

M. A. Bittinger, E. M. Driggers, V. R. Fantin, H. G. Jang,

S. Jin, M. C. Keenan, K. M. Marks, R. M. Prins, P. S. Ward,

K. E. Yen, L. M. Liau, J. D. Rabinowitz, L. C. Cantley,

C. B. Thompson, M. G. Vander Heiden and S. M. Su,

Nature, 2010, 465, 966; (b) J.-A. Losman and W. G. Kaelin

Jr, Genes Dev., 2013, 27, 836–852.

4 L. Dang and S.-S. M. Su, Annu. Rev. Biochem., 2017, 86, 305–

331.

5 (a) M. Tonjes, S. Barbus, Y. J. Park, W. Wang, M. Schlotter,

¨

A. M. Lindroth, S. V. Pleier, A. H. C. Bai, D. Karra,

R. M. Piro, J. Felsberg, A. Addington, D. Lemke,

I. Weibrecht, V. Hovestadt, C. G. Rolli, B. Campos,

S. Turcan, D. Sturm, H. Witt, T. A. Chan, C. Herold-Mende,

Data availability

Crystal structure data for His₆-AspH_315–758complexed to Mn,

2OG or a 2OG derivative (3-methyl-2OG, 16; 3-ethyl-2OG, 17; 4,4-

dimethyl-2OG, 34; N-oxalyl-a-methylalanine, 49), and, in some

cases, substrate peptide (hFX-EGFD1_86–124-4Ser) are deposited

in the protein data bank with PDB accession codes: 6YYU

(AspH:2OG), 6YYW (AspH:2OG:hFX-EGFD1_86–124-4Ser), 6YYV

(AspH:16), 6YYX (AspH:16:hFX-EGFD1_86–124-4Ser), 6YYY

(AspH:34:hFX-EGFD1_86–124-4Ser), 6Z6Q (AspH:17:hFX-EGFD1_86–

₁₂₄-4Ser), and 6Z6R (AspH:49:hFX-EGFD1_86–124-4Ser).

¨

R. Kemkemer, R. Konig, K. Schmidt, W.-E. Hull,

S. M. Pster, M. Jugold, S. M. Hutson, C. Plass, J. G. Okun,

G. Reifenberger, P. Lichter and B. Radlwimmer, Nat. Med.,

2013, 19, 901–908; (b) S. Raﬀel, M. Falcone, N. Kneisel,

J. Hansson, W. Wang, C. Lutz, L. Bullinger, G. Poschet,

Y. Nonnenmacher, A. Barnert, C. Bahr, P. Zeisberger,

¨

A. Przybylla, M. Sohn, M. Tonjes, A. Erez, L. Adler,

¨

P. Jensen, C. Scholl, S. Frohling, S. Cocciardi, P. Wuchter,

¨

C. Thiede, A. Florcken, J. Westermann, G. Ehninger,

P. Lichter, K. Hiller, R. Hell, C. Herrmann, A. D. Ho,

J. Krijgsveld, B. Radlwimmer and A. Trumpp, Nature, 2017,

551, 384–388.

Author contributions

L. B. synthesized the 2OG derivatives and the AspH substrate,

produced recombinant human AspH, and performed AspH

assays and crystallizations. Y. N. solved and rened the AspH

crystal structures. All authors analyzed data. L. B. and C. J. S.

wrote the manuscript with help from Y. N.

6 (a) R. Chowdhury, K. K. Yeoh, Y.-M. Tian, L. Hillringhaus,

E. A. Bagg, N. R. Rose, I. K. H. Leung, X. S. Li,

E. C. Y. Woon, M. Yang, M. A. McDonough, O. N. King,

I. J. Clion, R. J. Klose, T. D. W. Claridge, P. J. Ratcliﬀe,

C. J. Schoeld and A. Kawamura, EMBO Rep., 2011, 12,

463–469; (b) M. E. Figueroa, O. Abdel-Wahab, C. Lu,

P. S. Ward, J. Patel, A. Shih, Y. Li, N. Bhagwat,

A. Vasanthakumar, H. F. Fernandez, M. S. Tallman, Z. Sun,

Conﬂicts of interest

There are no conicts to declare.

Chem. Sci., 2021, 12, 1327–1342 | 1339

View Article Online

Chemical Science

Edge Article

K. Wolniak, J. K. Peeters, W. Liu, S. E. Choe, V. R. Fantin,

E. I. Traboulsi and F. S. Alkuraya, Am. J. Hum. Genet., 2014,

94, 755–759; (b) H. H. Abarca Barriga, N. Caballero,

M. Trubnykova, M. d. C. Castro-Mujica, J. E. La Serna-

¨

E. Paietta, B. Lowenberg, J. D. Licht, L. A. Godley,

R. Delwel, P. J. M. Valk, C. B. Thompson, R. L. Levine and

A. Melnick, Cancer Cell, 2010, 18, 553–567.

´

Infantes, F. Vasquez and R. C. Hennekam, Am. J. Med.

7 2-Oxoglutarate-Dependent Oxygenases, ed. C. J. Schoeld and

R. P. Hausinger, The Royal Society of Chemistry, Cambridge,

UK, 2015.

Genet., Part A, 2018, 176, 2494–2500; (c) O. M. Siggs,

E. Souzeau and J. E. Craig, Ophthalmic Genet., 2019, 40, 12–

16.

8 (a) K. D. Rasmussen and K. Helin, Genes Dev., 2016, 30, 733– 21 (a) G. P. Elvidge, L. Glenny, R. J. Appelhoﬀ, P. J. Ratcliﬀe,

750; (b) L. Shen, C.-X. Song, C. He and Y. Zhang, Annu. Rev.

Biochem., 2014, 83, 585–614.

9 (a) E. L. Greer and Y. Shi, Nat. Rev. Genet., 2012, 13, 343–357;

(b) J. C. Black, C. Van Rechem and J. R. Whetstine, Mol. Cell,

2012, 48, 491–507.

10 F. M. Vaz and R. J. A. Wanders, Biochem. J., 2002, 361, 417–

429.

11 J. Myllyharju, Matrix Biol., 2003, 22, 15–24.

12 (a) W. G. Kaelin and P. J. Ratcliﬀe, Mol. Cell, 2008, 30, 393–

402; (b) C. J. Schoeld and P. J. Ratcliﬀe, Nat. Rev. Mol. Cell

Biol., 2004, 5, 343–354.

J. Ragoussis and J. M. Gleadle, J. Biol. Chem., 2006, 281,

15215–15226; (b) P. J. Pollard, C. Loenarz, D. R. Mole,

M. A. McDonough, J. M. Gleadle, C. J. Schoeld and

P. J. Ratcliﬀe, Biochem. J., 2008, 416, 387–394; (c) Y. Benita,

H. Kikuchi, A. D. Smith, M. Q. Zhang, D. C. Chung and

R. J. Xavier, Nucleic Acids Res., 2009, 37, 4587–4602.

22 I. Pfeﬀer, L. Brewitz, T. Krojer, S. A. Jensen, G. T. Kochan,

N. J. Kershaw, K. S. Hewitson, L. A. McNeill, H. Kramer,

¨

M. Munzel, R. J. Hopkinson, U. Oppermann,

P. A. Handford, M. A. McDonough and C. J. Schoeld, Nat.

Commun., 2019, 10, 4910.

13 F. Korioth, C. Gieﬀers and J. Frey, Gene, 1994, 150, 395–399. 23 L. Brewitz, A. Tumber and C. J. Schoeld, J. Biol. Chem., 2020,

14 S. Kal and L. Que, J. Biol. Inorg. Chem., 2017, 22, 339–365. 295, 7826–7838.

15 (a) J. Steno, E. Holme, S. Lindstedt, N. Chandramouli, 24 M. A. McDonough, L. A. McNeill, M. Tilliet, C. A. Papamicael,

¨

L. H. Huang, J. P. Tam and R. B. Merrield, Proc. Natl.

Acad. Sci. U. S. A., 1989, 86, 444–447; (b) R. S. Gronke,

Q.-Y. Chen, B. Banerji, K. S. Hewitson and C. J. Schoeld, J.

Am. Chem. Soc., 2005, 127, 7680–7681.

W. J. VanDusen, V. M. Garsky, J. W. Jacobs, M. K. Sardana, 25 M. Breski, D. Dey, S. Obringer, B. Sudhamalla and K. Islam,

A. M. Stern and P. A. Friedman, Proc. Natl. Acad. Sci. U. S.

A., 1989, 86, 3609–3613.

16 (a) L. Lavaissiere, S. Jia, M. Nishiyama, S. de la Monte,

J. Am. Chem. Soc., 2016, 138, 13505–13508.

26 R. S. Wehbie, N. S. Punekar and H. A. Lardy, Biochemistry,

1988, 27, 2222–2228.

A. M. Stern, J. R. Wands and P. A. Friedman, J. Clin. 27 K. M. Steward and J. S. Johnson, Org. Lett., 2011, 13, 2426–

Invest., 1996, 98, 1313–1323; (b) K. Wang, J. Liu, Z.-L. Yan, 2429.

J. Li, L.-H. Shi, W.-M. Cong, Y. Xia, Q.-F. Zou, T. Xi, 28 (a) M. Xian, S. Alaux, E. Sagot and T. Geﬄaut, J. Org. Chem.,

F. Shen, H.-Y. Wang and M.-C. Wu, Hepatology, 2010, 52,

164–173.

2007, 72, 7560–7566; (b) E. Sagot, D. S. Pickering, X. Pu,

M. Umberti, T. B. Stensbøl, B. Nielsen, M. Chapelet,

J. Bolte, T. Geﬄaut and L. Bunch, J. Med. Chem., 2008, 51,

17 (a) K. S. Palumbo, J. R. Wands, H. Safran, T. King,

R. I. Carlson and S. M. de la Monte, Pancreas, 2002, 25, 39–

44; (b) X. Dong, Q. Lin, A. Aihara, Y. Li, C.-K. Huang,

W. Chung, Q. Tang, X. Chen, R. Carlson, C. Nadolny,

G. Gabriel, M. Olsen and J. R. Wands, Oncotarget, 2014, 6,

1231–1248; (c) H. Guofang, X. Boran, B. Yanghui,

W. Chuanlin, R. Beibei, S. Bei and B. Xuewei, Bosnian J.

Basic Med. Sci., 2018, 18, 297–304.

˙

4093–4103; (c) Z. Assaf, A. P. Larsen, R. Venskutonyte,

L. Han, B. Abrahamsen, B. Nielsen, M. Gajhede,

J. S. Kastrup, A. A. Jensen, D. S. Pickering, K. Frydenvang,

T. Geﬄaut and L. Bunch, J. Med. Chem., 2013, 56, 1614–

1628; (d) V. Helaine, J. Rossi and J. Bolte, Tetrahedron Lett.,

1999, 40, 6577–6580.

¨

29 S. Alaux, M. Kusk, E. Sagot, J. Bolte, A. A. Jensen, H. Brauner-

18 (a) N. Ince, S. M. de la Monte and J. R. Wands, Cancer Res.,

2000, 60, 1261–1266; (b) Q. Zou, Y. Hou, H. Wang,

Osborne, T. Geﬄaut and L. Bunch, J. Med. Chem., 2005, 48,

7980–7992.

K. Wang, X. Xing, Y. Xia, X. Wan, J. Li, B. Jiao, J. Liu, 30 C. B. Grissom and W. W. Cleland, Biochim. Biophys. Acta,

A. Huang, D. Wu, H. Xiang, T. M. Pawlik, H. Wang, Protein Struct. Mol. Enzymol., 1987, 916, 437–445.

W. Y. Lau, Y. Wang and F. Shen, EBioMedicine, 2018, 31, 31 H. H. Wasserman and W.-B. Ho, J. Org. Chem., 1994, 59,

287–298. 4364–4366.

19 (a) J. E. Dinchuk, R. J. Focht, J. A. Kelley, N. L. Henderson, 32 S. Faure, A. A. Jensen, V. Maurat, X. Gu, E. Sagot, D. J. Aitken,

N. I. Zolotarjova, R. Wynn, N. T. Neﬀ, J. Link, R. M. Huber,

T. C. Burn, M. J. Rupar, M. R. Cunningham, B. H. Selling,

J. Bolte, T. Geﬄaut and L. Bunch, J. Med. Chem., 2006, 49,

6532–6538.

J. Ma, A. A. Stern, G. F. Hollis, R. B. Stein and 33 (a) L. Ju, A. R. Lippert and J. W. Bode, J. Am. Chem. Soc., 2008,

P. A. Friedman, J. Biol. Chem., 2002, 277, 12970–12977; (b)

A. Aihara, C.-K. Huang, M. J. Olsen, Q. Lin, W. Chung,

Q. Tang, X. Dong and J. R. Wands, Hepatology, 2014, 60,

1302–1313.

130, 4253–4255; (b) S. P. Roche, S. Faure, L. El Blidi and

D. J. Aitken, Eur. J. Org. Chem., 2008, 2008, 5067–5078; (c)

M.-K. Wong, C.-W. Yu, W.-H. Yuen and D. Yang, J. Org.

Chem., 2001, 66, 3606–3609.

20 (a) N. Patel, A. O. Khan, A. Mansour, J. Y. Mohamed, A. Al- 34 V. R. Pattabiraman, A. O. Ogunkoya and J. W. Bode, Angew.

´

Assiri, R. Haddad, X. Jia, Y. Xiong, A. Megarbane,

Chem., Int. Ed., 2012, 51, 5114–5118.

1340 | Chem. Sci., 2021, 12, 1327–1342

View Article Online

Edge Article

Chemical Science

35 F.-P. Wu, J.-B. Peng, L.-S. Meng, X. Qi and X.-F. Wu,

ChemCatChem, 2017, 9, 3121–3124.

36 R. F. Heck, J. Am. Chem. Soc., 1968, 90, 5518–5526.

Biochemistry, 1982, 21, 3690–3697; (e) G. E. Smyth and

R. F. Colman, Arch. Biochem. Biophys., 1992, 293, 356–361;

(f) R. S. Ehrlich and R. F. Colman, J. Biol. Chem., 1987, 262,

12614–12619; (g) R. A. Bednar and R. F. Colman, J. Biol.

Chem., 1982, 257, 11734–11739; (h) R. S. Ehrlich and

R. F. Colman, J. Biol. Chem., 1984, 259, 11936–11942.

¨

37 C. Gurtler and S. L. Buchwald, Chem. Eur. J., 1999, 5, 3107–

3112.

38 (a) L. Horner, H. Hoﬀmann and H. G. Wippel, Chem. Ber.,

´

1958, 91, 61–63; (b) W. S. Wadsworth and W. D. Emmons, 53 (a) H. Chen, S. Laurent, S. Bedu, F. Ziarelli, H.-l. Chen,

J. Am. Chem. Soc., 1961, 83, 1733–1738.

39 H. Wissmann and H.-J. Kleiner, Angew. Chem., Int. Ed., 1980,

19, 133–134.

40 J.-H. Zhang, T. D. Y. Chung and K. R. Oldenburg, J. Biomol.

Screening, 1999, 4, 67–73.

41 (a) L. Brewitz, A. Tumber, I. Pfeﬀer, M. A. McDonough and

C. J. Schoeld, Sci. Rep., 2020, 10, 8650; (b) L. Brewitz,

A. Tumber, A. Thalhammer, E. Salah, K. E. Christensen

and C. J. Schoeld, ChemMedChem, 2020, 15, 1139–1149;

(c) L. Brewitz, A. Tumber, X. Zhang and C. J. Schoeld,

Bioorg. Med. Chem., 2020, 28, 115675.

Y. Cheng, C.-C. Zhang and L. Peng, Chem. Biol., 2006, 13,

849–856; (b) Y. Wang, X. Liu, E. Laurini, P. Posocco,

F. Ziarelli, M. Fermeglia, F. Qu, S. Pricl, C.-C. Zhang and

L. Peng, Org. Biomol. Chem., 2014, 12, 4723–4729; (c)

X. Liu, Y. Wang, E. Laurini, P. Posocco, H. Chen,

F. Ziarelli, A. Janicki, F. Qu, M. Fermeglia, S. Pricl,

C.-C. Zhang and L. Peng, Org. Lett., 2013, 15, 4662–4665;

(d) M. Fang, A. Macova, K. L. Hanson, J. Kos and

D. R. J. Palmer, Biochemistry, 2011, 50, 8712–8721; (e)

V. I. Bunik, T. T. Denton, H. Xu, C. M. Thompson,

A. J. L. Cooper and G. E. Gibson, Biochemistry, 2005, 44,

10552–10561; (f) A. V. Artiukhov, A. Grabarska,

42 A. V. Hill, J. Physiol., 1910, 40, iv–vii.

¨

43 B. K. Shoichet, J. Med. Chem., 2006, 49, 7274–7277.

E. Gumbarewicz, V. A. Aleshin, T. Kahne, T. Obata,

¨

44 P. Koivunen, M. Hirsila, A. M. Remes, I. E. Hassinen,

A. V. Kazantsev, N. V. Lukashev, A. Stepulak, A. R. Fernie

and V. I. Bunik, Sci. Rep., 2020, 10, 1886.

54 C. P. Ashcro, S. Challenger, A. M. Derrick, R. Storey and

N. M. Thomson, Org. Process Res. Dev., 2003, 7, 362–368.

K. I. Kivirikko and J. Myllyharju, J. Biol. Chem., 2007, 282,

4524–4532.

45 R. S. Gronke, D. J. Welsch, W. J. VanDusen, V. M. Garsky,

M. K. Sardana, A. M. Stern and P. A. Friedman, J. Biol. 55 (a) T. T. Zengeya, R. A. Kulkarni and J. L. Meier, Org. Lett.,

Chem., 1990, 265, 8558–8565.

46 (a) E. A. Siess, D. G. Brocks, H. K. Lattke and O. H. Wieland,

Biochem. J., 1977, 166, 225–235; (b) K. Thirstrup,

2015, 17, 2326–2329; (b) W. Xu, H. Yang, Y. Liu, Y. Yang,

P. Wang, S.-H. Kim, S. Ito, C. Yang, P. Wang, M.-T. Xiao,

L.-x. Liu, W.-q. Jiang, J. Liu, J.-y. Zhang, B. Wang, S. Frye,

Y. Zhang, Y.-h. Xu, Q.-y. Lei, K.-L. Guan, S.-m. Zhao and

Y. Xiong, Cancer Cell, 2011, 19, 17–30.

´

¨

S. Christensen, H. A. Møller, A. Ritzen, A.-L. Bergstrom,

T. N. Sager and H. S. Jensen, Pharmacol. Res., 2011, 64,

268–273.

56 K. Al-Qahtani, B. Jabeen, R. Sekirnik, N. Riaz,

T. D. W. Claridge, C. J. Schoeld and J. S. O. McCullagh,

Phytochemistry, 2015, 117, 456–461.

47 L. Ingraham, M. Li, J. L. Renfro, S. Parker, A. Vapurcuyan,

I. Hanna and R. M. Pelis, Mol. Pharmacol., 2014, 86, 86–95.

˚

48 (a) K. Valegard, A. C. T. van Scheltinga, M. D. Lloyd, T. Hara, 57 (a) R. R. Raval, K. W. Lau, M. G. B. Tran, H. M. Sowter,

S. Ramaswamy, A. Perrakis, A. Thompson, H.-J. Lee,

J. E. Baldwin, C. J. Schoeld, J. Hajdu and I. Andersson,

Nature, 1998, 394, 805–809; (b) M. A. McDonough,

K. L. Kavanagh, D. Butler, T. Searls, U. Oppermann and

C. J. Schoeld, J. Biol. Chem., 2005, 280, 41101–41110.

S. J. Mandriota, J.-L. Li, C. W. Pugh, P. H. Maxwell,

A. L. Harris and P. J. Ratcliﬀe, Mol. Cell. Biol., 2005, 25,

5675; (b) A. Loboda, A. Jozkowicz and J. Dulak, Mol. Cells,

2010, 29, 435–442.

´

58 R. M. Labbe, A. Holowatyj and Z.-Q. Yang, Am. J. Transl. Res.,

49 (a) P. Fernlund and J. Steno, J. Biol. Chem., 1983, 258,

2014, 6, 1–15.

12509–12512; (b) B. A. McMullen, K. Fujikawa, W. Kisiel, 59 (a) S. Hamada, T.-D. Kim, T. Suzuki, Y. Itoh, H. Tsumoto,

T. Sasagawa, W. N. Howald, E. Y. Kwa and B. Weinstein,

Biochemistry, 1983, 22, 2875–2884.

H. Nakagawa, R. Janknecht and N. Miyata, Bioorg. Med.

Chem. Lett., 2009, 19, 2852–2855; (b) N. R. Rose, S. S. Ng,

´

50 (a) A. J. L. Cooper and A. G. Redeld, J. Biol. Chem., 1975, 250,

527–532; (b) J. T. Smith and V. M. Doctor, J. Inorg. Nucl.

Chem., 1975, 37, 775–777; (c) X. Yang, Q. Zhang, L. Li and

R. Shen, J. Inorg. Biochem., 2007, 101, 1242–1250.

J. Mecinovic, B. M. R. Lienard, S. H. Bello, Z. Sun,

M. A. McDonough, U. Oppermann and C. J. Schoeld, J.

Med. Chem., 2008, 51, 7053–7056; (c) N. R. Rose,

´

E. C. Y. Woon, G. L. Kingham, O. N. F. King, J. Mecinovic,

51 (a) G. W. E. Plaut, T. Aogaichi and J. L. Gabriel, Arch.

Biochem. Biophys., 1986, 245, 114–124; (b) T. T. Denton,

C. M. Thompson and A. J. L. Cooper, Anal. Biochem., 2001,

I. J. Clion, S. S. Ng, J. Talib-Hardy, U. Oppermann,

M. A. McDonough and C. J. Schoeld, J. Med. Chem., 2010,

53, 1810–1818.

298, 265–274; (c) Y. Kato, Y. Asano, T. K. Makar and 60 B. Sudhamalla, S. Wang, V. Snyder, S. Kavoosi, S. Arora and

A. J. L. Cooper, J. Biochem., 1996, 120, 531–539. K. Islam, J. Am. Chem. Soc., 2018, 140, 10263–10269.

52 (a) R. A. Bednar, F. C. Hartman and R. F. Colman, 61 V. Helaine, J. Rossi, T. Geﬄaut, S. Alaux and J. Bolte, Adv.

Biochemistry, 1982, 21, 3681–3689; (b) F. C. Hartman, Synth. Catal., 2001, 343, 692–697.

Biochemistry, 1981, 20, 894–898; (c) A. Saha, Y.-C. Huang 62 (a) K. Majamaa, H. M. Hanauske-Abel, V. Gunzler and

´

¨

and R. F. Colman, Biochemistry, 1989, 28, 8425–8431; (d)

K. I. Kivirikko, Eur. J. Biochem., 1984, 138, 239–245; (b)

R. A. Bednar, F. C. Hartman and R. F. Colman,

Chem. Sci., 2021, 12, 1327–1342 | 1341

View Article Online

Chemical Science

Edge Article

C. J. Cunliﬀe, T. J. Franklin, N. J. Hales and G. B. Hill, J. Med. 70 T. Hamada, S. Matsunaga, G. Yano and N. Fusetani, J. Am.

Chem., 1992, 35, 2652–2658. Chem. Soc., 2005, 127, 110–118.

63 M. Mukherji, N. J. Kershaw, C. H. MacKinnon, I. J. Clion, 71 L. Robbel and M. A. Marahiel, J. Biol. Chem., 2010, 285,

A. S. Wierzbicki, C. J. Schoeld and M. D. Lloyd, Chem.

Commun., 2001, 972–973.

64 (a) S. Martinez and R. P. Hausinger, Biochemistry, 2016, 55,

5989–5999; (b) S. Martinez, M. Fellner, C. Q. Herr,

A. Ritchie, J. Hu and R. P. Hausinger, J. Am. Chem. Soc.,

2017, 139, 11980–11988.

65 G. R. Moran, Arch. Biochem. Biophys., 2014, 544, 58–68.

66 P. O. Larsen and E. Wieczorkowska, Biochim. Biophys. Acta,

Gen. Subj., 1978, 542, 253–262.

27501–27508.

72 (a) C. Milne, A. Powell, J. Jim, M. Al Nakeeb, C. P. Smith and

J. Mickleeld, J. Am. Chem. Soc., 2006, 128, 11250–11259; (b)

C. Mahlert, F. Kopp, J. Thirlway, J. Mickleeld and

M. A. Marahiel, J. Am. Chem. Soc., 2007, 129, 12011–12018.

73 (a) D. Mukherjee and M. M. Laloraya, Biochem. Physiol.

Panz., 1974, 166, 429–436; (b) A. I. Virtanen and

A.-M. Berg, Acta Chem. Scand., 1955, 9, 553–554.

´

˜

74 R. Dıaz, H. Gallart-Ayala, J. V. Sancho, O. Nunez, T. Zamora,

´

67 A. L. Wilkins, Y. Lu and S.-T. Tan, J. Agric. Food Chem., 1995,

43, 3021–3025.

C. P. B. Martins, F. Hernandez, S. Hernandez-Cassou,

J. Saurina and A. Checa, J. Chromatogr. A, 2016, 1433, 90–97.

68 (a) F. Guneral and C. Bachmann, Clin. Chem., 1994, 40, 862– 75 M. Brauchle, Z. Yao, R. Arora, S. Thigale, I. Clay, B. Inverardi,

´

868; (b) M. T. Garcıa-Silva, A. Ribes, Y. Campos, B. Garavaglia

J. Fletcher, P. Taslimi, M. G. Acker, B. Gerrits, J. Voshol,

¨

and J. Arenas, Pediatr. Neurol., 1997, 17, 165–170.

A. Bauer, D. Schubeler, T. Bouwmeester and H. Ruﬀner,

69 (a) M. J. M. Nowaczyk, D. C. Lehotay, B.-A. Platt, L. Fisher,

PLoS One, 2013, 8, e60549.

R. Tan, H. Phillips and J. T. R. Clarke, Metabolism, 1998, 76 J. R. Horton, A. K. Upadhyay, H. Hashimoto, X. Zhang and

47, 836–839; (b) V. Tiranti, E. Briem, E. Lamantea, X. Cheng, J. Mol. Biol., 2011, 406, 1–8.

R. Mineri, E. Papaleo, L. De Gioia, F. Forlani, P. Rinaldo, 77 L. Liu, H. Kim, A. Casta, Y. Kobayashi, L. S. Shapiro and

P. Dickson, B. Abu-Libdeh, L. Cindro-Heberle, M. Owaidha,

R. M. Jack, E. Christensen, A. Burlina and M. Zeviani, J.

Med. Genet., 2006, 43, 340–346.

A. M. Christiano, FASEB J., 2014, 28, 1534–1542.

1342 | Chem. Sci., 2021, 12, 1327–1342

Article Doi

Synthesis of 2-oxoglutarate derivatives and their evaluation as cosubstrates and inhibitors of human aspartate/asparagine-β-hydroxylase

DOI: 10.1039/d0sc04301j

Source and publish data:

Authors:

Article abstract of DOI:10.1039/d0sc04301j

Full text of DOI:10.1039/d0sc04301j

Products guided by the article

R&D Labs maybe for 4436-03-7

Relevant to this article

Hot Product