Antimalarial activity of primaquine operates via a two-step biochemical relay

Article abstract of DOI:10.1038/s41467-019-11239-0

Primaquine (PQ) is an essential antimalarial drug but despite being developed over 70 years ago, its mode of action is unclear. Here, we demonstrate that hydroxylated-PQ metabolites (OH-PQm) are responsible for efficacy against liver and sexual transmission stages of Plasmodium falciparum. The antimalarial activity of PQ against liver stages depends on host CYP2D6 status, whilst OH-PQm display direct, CYP2D6-independent, activity. PQ requires hepatic metabolism to exert activity against gametocyte stages. OH-PQm exert modest antimalarial efficacy against parasite gametocytes; however, potency is enhanced ca.1000 fold in the presence of cytochrome P450 NADPH:oxidoreductase (CPR) from the liver and bone marrow. Enhancement of OH-PQm efficacy is due to the direct reduction of quinoneimine metabolites by CPR with the concomitant and excessive generation of H₂O₂, leading to parasite killing. This detailed understanding of the mechanism paves the way to rationally re-designed 8-aminoquinolines with improved pharmacological profiles.

Full text of DOI:10.1038/s41467-019-11239-0

A
R
T
I
C
L
E
https://doi.org/10.1038/s41467-019-11239-0
OPEN
A
n
a
t
a
i
m
a
w
l
a
o
r
-
i
s
a
t
l
e
a
p
c
t
b
i
i
v
o
i
t
c
y
h
o
f
p
r
a
i
m
a
e
q
u
i
n
e
o
p
e
r
a
t
e
s
v
i
t
e
m
i
c
l
r
l
a
y
1
1
1
,
7
1
,
8
2
G
r
a
z
i
a
C
a
m
a
r
d
a
,
P i
y
a
p
o
r
n
J
i
r
a
w a
t
c
h
a
r
a
d
e c h
,
R
i
c
h
a
r
d
S
.
P
r
i
e
s
t
l
e
y
,
A
h
m
e
d
S
a
i
f
,
S
a
n
d
r
a
M
a
r
ch
,
3
3
2
4
5
6
M
i
c
h
a
e
l
H
.
L
.
W
o
n
g
,
S
ue
t
L
e
u
ng
,
A
le
x
B
.
M
i
l
le
r
,
D
a
v
i
d
A
.
B
a
ke
r
,
P
i
e
tr
o
A
l
a
n
o
,
M
a
r
k
J
.
I
.
P
a
in
e
,
2
3
1
1
S
a
n
g
e
e
t
a
N
.
B
h
a
ti
a
,
P
a
u
l
M
.
O ’N
e
i
l
l
,
S
t
e
p
h
e
n
A
.
W
a
r
d
&
G
i
a
n
c
a
r
l
o
A
.
B
i
a
g
i
n
i
P
r
i
m
a
q
u
i
n
e
o
(
P
Q
)
i
s
a
n
e
s
s
e
n
t
i
a
l
a
n
t
i
m
a
l
a
r
i
a
l
d
r
u
g
b
u
t
d
e
s
p
i
t
e
b
t
e
i
n
g
d
e
v
e
l
o
p
e
d
o
v
e
r
7
0
y
e
a
r
s
a
g
o
,
i
t
s
m
d
e
o
e
f
a
c
t
i
o
n
i
s
u
n
c
l
e
a
r
.
H
e
r
e
,
w
e
d
e
m
o
n
s
t
r
a
t
e
t
h
a
h
y
d
r
o
x
y
l
a
t
e
d
-
P
Q
m
e
t
a
b
o
l
i
t
e
s
(
O
H
-
P
Q
m
)
a
r
r
e
s
p
o
n
s
i
b
l
e
f
o
r
e
f
fi
c
a
c
y
a
g
a
i
n
s
t
l
i
v
e
r
a
n
d
s
e
x
u
a
l
t
r
a
n
s
m
i
s
s
i
o
n
s
t
a
g
e
s
o
f
Plasmodium falciparum
.
T
h
e
a
n
t
i
m
a
l
a
r
i
a
l
a
c
t
i
v
i
t
y
o
f
P
Q
a
g
a
i
n
s
t
l
i
v
e
r
s
t
a
g
e
s
d
e
t
p
e
n
d
s
o
n
h
o
s
t
C
Y
P
2
D
6
s
t
a
t
u
s
,
w
h
i
l
s
t
o
O
H
-
P
Q
m
d
i
s
p
l
a
y
d
i
r
e
c
t
,
C
Y
P
2
D
6
-
i
n
d
e
p
p
e
n
d
e
n
t
,
a
c
t
i
v
i
y
.
P
Q
r
e
q
u
i
r
e
s
h
e
p
a
t
i
c
m
e
t
a
b
o
l
i
s
m
t
e
x
e
r
t
a
c
t
i
v
i
t
y
a
g
a
i
n
s
t
g
a
m
e
t
o
c
y
t
e
s
t
a
g
e
s
.
O
H
-
P
Q
m
e
x
e
r
t
m
o
d
e
s
t
a
n
t
i
m
a
l
a
r
i
a
l
e
f
fi
c
a
c
y
e
a
g
f
a
i
n
s
t
p
h
a
r
a
s
i
t
e
g
a
m
e
t
o
c
y
t
e
s
;
h
o
w
e
v
e
r
,
o
t
e
n
c
y
i
s
e
n
h
a
n
c
e
d
c
a
.
1
0
0
0
f
o
l
d
n
i
n
t
h
e
p
r
e
s
e
n
c
o
c
y
t
o
c
r
o
m
e
P
4
5
0
N
A
D
P
H
:
o
x
i
d
o
r
e
d
u
c
t
a
s
e
(
C
P
R
)
f
r
o
m
t
h
e
l
i
v
e
r
a
n
d
b
n
o
e
m
a
r
r
o
w
.
E
n
h
a
n
c
e
m
C
e
n
t
o
f
O
H
-
P
Q
m
e
f
fi
c
a
c
y
i
s
d
u
e
t
o
t
h
e
d
i
re
e
c
t
r
e
d
u
c
t
i
o
n
o
f
q
u
i
n
o
-
e
i
m
i
n
e
m
e
t
a
b
o
l
i
t
e
s
b
y
P
R
w
i
t
h
t
h
e
c
o
n
c
o
m
i
t
a
n
t
a
n
d
e
x
c
e
s
s
i
v
e
g
e
n
r
a
t
i
o
n
o
f
H
y
O
,
l
e
a
d
i
n
g
2
2
t
o
p
a
r
a
s
i
t
e
k
i
l
l
in
g
.
T
h
i
s
d
e
t
a
i
l
e
d
u
n
d
e
r
s
t
a
n
d
i
n
g
o
f
t
h
e
m
e
c
h
a
n
i
s
m
p
a
v
e
s
t
h
e
w
a
t
o
r
a
t
i
o
n
a
l
l
y
r
e
-
d
e
s
ig
n
ed
8
-a
m
in
o
q
u
i
n
o
l
i
n
e
s
w
i
th
i
m
p
r
o
v
e
d
p
h
a
r
m
a
c
o
l
o
g
i
c
a
l
p
ro fi
l
e
s
.
1 Centre for Drugs and Diagnostics Research, Tropical Disease Biology Department, Liverpool School of Tropical Medicine, Liverpool L3 5QA, UK. 2 Health
Sciences and Technology/Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA 02139, USA. 3 Department
of Chemistry, University of Liverpool, Liverpool L69 7ZD, UK. 4 Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine,
London WC1E 7HT, UK. 5 Dipartimento di Malattie Infettive, Istituto Superiore di Sanità, Rome 00161, Italy. 6 Vector Biology Department, Liverpool School of
Tropical Medicine, Liverpool L3 5QA, UK. 7Present address: ARUK Oxford Drug Discovery Institute, University of Oxford, Oxford OX3 7FZ, UK. 8Present
address: Clinical Laboratory sciences Department, College of Applied Medical Sciences, Najran University, Najran 61441, Saudi Arabia. Correspondence and
requests for materials should be addressed to G.A.B. (email: Giancarlo.Biagini@lstmed.ac.uk)
N
A
T
U
R
E
C
OM
M
U
N
I
C
A
T
I
ON
S
|
(2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
ww
.n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
re
c
o
m
m
u
n
i
c
a
t
i
o
n
s
1

A
R
T
I
C
L
E
N A T U R E C O M MU N ICA T IO N S | h t t p s : / /d o i .o r g / 1 0 .1 0 38 / s 41 46 7 - 01 9 -1 1 2 39 - 0
ince its development in the 1950s by the US Army, an despite intensive research efforts the exact enzyme(s) and meta-
estimated 200 million people have been administered PQ1. bolic pathway(s) involved in the process remain to be fully
Research into the elusive PQ mode of action has mainly elucidated.
S
focused on the identification of the biochemical basis for its side
Similarly, the mechanism of action of primaquine against the
effect of hemolytic toxicity in patients with G6PD deficiency1. malaria parasite is also largely unknown6,7. A link between drug
Earlier studies led to the identification of OH-PQm, generated efficacy and metabolism through CYP2D6 is supported by recent
from hepatic phase I metabolism2, as potential culprits for animal and clinical studies, including the association of CYP2D6
toxicity3,4. The most relevant species in this context include the poor metaboliser phenotype status with primaquine failure in
5-hydroxy (5-HPQ) and 5,6-dihydroxy (5,6-DPQ) primaquine controlled human malaria infections with P. vivax2,8. The need
metabolites (Fig. 1a). These species are unstable, undergoing for CYP2D6 metabolism to generate key OH-PQm9,10 supports
spontaneous oxidation and producing the corresponding quino- the assumption that these metabolites contribute to drug efficacy;
neimine forms (Fig. 1a, PQQI, 5-quinoneimine, and 6OHPQQI, however, hitherto there has been no direct demonstration of the
6-hydroxy-5-quinoneimine) with concomitant generation of link between these metabolites and anti-parasitic activity. Any
5
H2O2 ; quinoneimine metabolites can then be reduced back to explanation of PQ action needs to address the exquisite selectivity
the hydroxylated form, leading to H2O2 accumulation. However, against dormant and active liver stage parasites and the ability of
their reduction potential is likely to be very negative, limiting the very low drug doses and subsequent low systemic exposures to
pool of possible intracellular reductants with a sufficiently low kill gametocyte stages of P. falciparum clinically11. This study
reduction potential to donate electrons. In proof-of-concept attempts to reconcile these questions.
studies using the strong reducing enzyme, ferredoxin-NADP
Here OH-PQm in their hydroxyquinoline or quinoneimine
reductase (FNR) from spinach (note the Em,7 known for form have been synthesised and assessed for their activity against
the ferredoxin oxidised/reduced couple is ca. −430 mV), PQ P. falciparum gametocytes and liver stages. By assessing parasite
quinoneimines were shown to be enzymatically reduced5—but viability in the presence or absence of metabolic conversion we
a
NH2
NH2
NH2
NH2
NH2
NH
NH
N
NH
N
N
N
N
N
N
O
O
O
HO
HO
OH
O
OH
O
b
c
d
e
f
YEM, IC50 0.558 μM
NON, IC50 0.772 μM
YEM, IC50 0.522 μM
NON, IC50 0.494 μM
YEM, IC50 0.172 μM
NON, IC50 0.464 μM
YEM, IC50 5.873 μM
NON, IC50 0.404 μM
YEM, IC50 0.611 μM
NON, IC50 1.274 μM
150
150
150
150
150
100
50
100
50
0
100
50
0
100
50
0
100
50
0
0
0.001 0.01 0.1
[PQ], μM
1
10 100
0.001 0.01 0.1
1
10
0.001 0.01 0.1
1
10
0.001 0.01 0.1
[5,6-DPQ], μM
1
10
0.001 0.01 0.1
1
10
[5-HPQ], μM
[PQQI], μM
[6OHPQQI], μM
g
h
i
(–)2D6
(+)2D6
No CPR
huCPR
huCPR + pyruvate
(–)HLM
(+)HLM
(+)2D6 + paroxetine
huCPR-No NADP
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
PQ
PQ
PQQI
PQQI
PQQI
5-HPQ
5-HPQ
5-HPQ
5,6-DPQ
5,6-DPQ
5,6-DPQ
6OHPQQI
6OHPQQI
6OHPQQI
Fig. 1 S t ru ct u re s o f P Q a n d O H- P Qm a n d t h e i r a c t i vi t y a g a i n s t P. falciparum l i v e r s t a g e s a n d g a m e t o c y t e s . a S t r u c t u re s o f c o m p o u n d s u s e d i n t h is s t u dy .
b, f D o s e d e p e n d e n t r ed u c t i o n i n e x o e r y t h r o cy t i c f o r m s ( E EF s) n u m b e rs a t d a y 3 . 5 p o s t i n f e c t i o n w i t h N F 5 4 s p o r o z o i te s i n Y EM ( p o o r m e t a b ol i s er , b la c k
l i n e ) a n d N O N ( ex t e n s i v e m e t a b o li s er , d a s h e d b l u e l in e) h e p a t o c y t e l o t s . b P r i m a q u i n e ( P Q) . c 5 - h y d r o x y- p r i m a q u i ne ( 5 - H P Q) . d 5 - q u i n o n e i mi ne ( P Q Q I ) .
e 5 , 6 - d i h y d r ox y- p r i m a q u i n e ( 5 , 6 -D P Q) . f 6 - h y d r o xy - 5 - q ui n o n e i m i n e ( 6 O H P Q QI) . V i a b i l i t y i s e x p r e s se d a s m e a n p e r c e n t a g e o f v e h i c l e c o n t r o l S . D .
o
f
t
w
o
i
n
d
e
p
e
n
d
e
n
t
e
x
p
e
r
i
m
e
n
t
s
p
e
r
f
o
r
m
e
d
i
n
t
r
i
p
l
i
c
a
t
e
s
.
F
o
r
e
a
c
h
d
r
u
g
,
I
C
v
a
l
u
e
s
a
r
e
r
e
p
o
r
t
e
d
f
o
r
t
h
e
p
o
o
r
m
e
t
a
b
o
l
i
s
e
r
,
Y
E
M
,
a
n
d
e
x
t
e
n
s
i
v
e
m
e
t
a
b
o
l
i
s
e
r
,
5
0
N O N , h e p a to cyt e l o t s . g C o u p l ed i n v i tr o m e t a b o li sm - G C -L U C a s s a y . P Q a n d P Q m e t a b o l i te s ( 3 0 μM ) w e r e r e a ct e d w i t h ( b l u e s q u a r e s) o r w i t ho ut
( r e d c i r cl e s h u m a n l i ve r m i c r o s o m e s ( H L M ) p r i o r t o d i l u t io n t o 1 0 μM ( n o m i n a l p a r e n t a l c o m p o u n d c o n c e n t ra t i o n ) i n a G C - L U C a s s a y w i t h m a t ur e
g a m e t o c yt es . V i a b i l i t y w a s m e a s u r e d a f t e r 7 2 h a n d e x p r e s s ed a s m e a n p e r c e n t a g e o f c o n t r o l ( n o d r u g) v i a b i l i t y S . D . ( n = 3 , e a c h i n t r i p l i ca t e ). h a s i n
g, b u t w i th ( b l u e s q u a re s) o r w it h o u t ( r e d c i r c l e s ) C Y P2 D 6 ( n = 2 , 5 t o t a l r e p l i c a t e s) . F o r p a r o x e t i n e i n h i b i t i o n , C Y P 2 D6 w a s p r e- i n c u b a t ed w i t h 1 0 μM
p a ro x e t i n e f o r 1 5 m i n p r i o r t o c o m p o u n d s ( b la c k t r ia n g le s , n = 2 , 4 t o t a l r e p l i c a te s ) ; i a s i n g, b u t w i t h o u t ( r e d c i r cl es ) o r w i t h ( b l u e s q u a r e s) r e c o m bi nan t
h u m a n C P R, o r w i th h u C P R i n t h e p r e s e n c e o f 1 0 m M s o d i u m p y r u v a te ( b r o w n d i a m o n d s) w i t h h u C P R i n t h e a b s en ce o f N A P D + ( b l a c k t r i a n g l e s) . N oN A D P
+
n = 1 i n d u p l i c a t e; a l l o t h e r h u C P R c o n d i t io n s n = 4 i n d u p li c a te
2
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 2 3 9 - 0
A R T I C L E
show that OH-PQm full cidal potential requires further enzy- gametocytocidal activity significantly increased, confirming that
matic catalysis. We identify the CYP2D6 redox partner, cyto- metabolic transformation is also required for gametocyte killing
chrome P450 NADPH:oxidoreductase (CPR), as required for activity. Interestingly, a significant increase in potency was also
OH-PQm to exert full gametocytocidal activity, independently of observed for all OH-PQm after microsomal incubation (p value <
CYP2D6 status. We also show that OH-PQm activity against 0.001 for all pairs by Mann Whitney test, Supplementary
P. falciparum liver stage development is comparable or better Table 1). In both the presence and absence of metabolic con-
than PQ and confirm a dispensable role for CYP2D6 activity. version, 5,6-DPQ showed the highest gametocytocidal potency
Finally, we substantiate the critical role of H2O2 production in (Supplementary Table 2). These results confirm that PQ requires
anti-parasitic activity and propose a CPR-mediated biochemical metabolic activation to elicit activity against both liver stage
model to account for PQ and OH-PQm mechanism of action. parasites and gametocytes; gametocytocidal activity of OH-PQm
These data answer long-held questions of why PQ displays in the in vitro gametocyte setting is also shown to be HLM
parasite stage-specific activity and why PQ displays transmission metabolism-dependent.
blocking activity at such low doses.
To gain further insight, we directly assessed the role of
CYP2D6 in OH-PQm activity against gametocytes by performing
coupled in vitro metabolism-GC-LUC assays using CYP2D6-
expressing baculosomes (Fig. 1h). As with HLM, CYP2D6
treatment potentiated OH-PQm gametocytocidal activity, with
p < 0.05 for 5,6-DPQ and 6OHPQQI, and p < 0.001 for all other
compounds (Mann Whitney, Supplementary Table 3). However,
OH-PQm are believed to be the terminal products of primaquine
metabolism, so we further investigated these results by blocking
CYP2D6 activity with the specific inhibitor paroxetine15
(Supplementary Fig. 2) prior to drug in vitro metabolism and
GC-LUC assays. Interestingly, CYP2D6 inhibition did not
significantly affect OH-PQm gametocytocidal activity (Fig. 1h,
compare (+)2D6 and (+)2D6 + Paroxetine); conversely, PQ
activity decreased (p < 0.05) as expected, reverting towards the
levels observed in control samples without CYP2D6 (Mann
Whitney, Supplementary Table 4). Overall, these results show
that OH-PQm activity is greatly enhanced by baculosome
metabolic component(s), however, the observed potentiation is
independent of CYP2D6 activity.
Results
Influence of CYP2D6 on PQ antimalarial liver-stage activity.
To explore the activity of PQ and OH-PQm against P. falciparum
liver stage development, we used an advanced in vitro model that
consistently mimics liver physiology and supports complete
maturation of P. falciparum extra-erythrocyte forms (EEFs) in
the liver12. In this in vitro system, human primary hepatocytes
are organised into colonies surrounded by supportive stromal
cells (Micropatterned coculture, MPCC) and infected with
P. falciparum (NF54 strain) sporozoites12. To evaluate the role of
CYP2D6 in PQ and OH-PQm activity, we selected two hepato-
cyte lots, NON and YEM. YEM is a low metaboliser based on
genotype analysis, it has a variation in the CYP2D6 allele (*4/*4);
this was confirmed by incubation with the CYP2D6 probe deb-
risoquine13 and measuring the metabolite 4-hydroxidebrisoquine
by mass spectrometry (MS) (Supplementary Fig. 1), confirming
that the NON hepatocyte lot had twofold higher CYP2D6 activity
than the YEM lot.
These results prompted us to hypothesise that OH-PQm might
be direct substrates of CPR, the CYPs redox partner required for
electron transfer from NAPDH16, an intrinsic (and required)
component of HLM and CYP2D6 baculosome preparations. We
therefore measured initial reaction rates of human recombinant
CPR (huCPR)17 with PQ and OH-PQm (Supplementary Fig. 5)
and generated the apparent steady-state kinetics for each
compound (Supplementary Table 5). These analyses revealed
that OH-PQm are indeed CPR substrates; conversely, kinetic
parameters showed parental PQ to be a poor CPR substrate, as
predictable from its structure, and hence it was not included in
the subsequent analysis. We then tested the gametocytocidal
ability of OH-PQm following incubation with huCPR and
observed that, akin to HLM and the baculosome treatment, in
the presence of huCPR only, the gametocytocidal activity of all
metabolites was significantly increased (Fig. 1i, p < 0.001 for all
compounds, Mann Whitney, Supplementary Table 6). This effect
was completely prevented when the assay was performed in the
OH-PQm were synthesised and used alongside PQ to directly
assess the inhibition of EEFs development in dose-response
experiments in the two hepatocyte lots. As shown in Fig. 1b, PQ
activity against EEFs development strictly depended on hepato-
cyte CYP2D6 status, with >14-fold increase in the IC50 measured
in the YEM background (low metaboliser, PQ IC50 5.87 μM) as
compared to NON hepatocytes (extensive metaboliser PQ IC50
0.40 μM). Conversely, all OH-PQm showed comparable IC50
values in the two hepatocyte lots (Fig. 1c–f), indicating that they
act independently of CYP2D6 activity. It is interesting to note
that all OH-PQm show similar potency against liver stages,
comparable to that seen for PQ in the presence of extensive
metaboliser hepatocytes. Taken together these results confirm
that OH-PQm have direct killing activity against liver stages,
supporting the hypothesis that they act downstream of CY2D6
activity.
PQ gametocytocidal activity requires hepatic metabolism. absence of NADP+, required for the production of the essential
We next explored the anti-gametocyte activity of OH-PQm cofactor NADPH by the reaction components (see “Methods”)
and parental PQ through coupled in vitro metabolism- (Fig. 1i, compare (+)huCPR and (+)huCPR-No NADP+),
gametocytocidal assays. In these assays, we first incubated PQ confirming that huCPR catalysis mediates OH-PQm gametocy-
and OH-PQm with human liver microsomes (HLM) to mimic tocidal activity. Cognisant of the putative role of H2O2-producing
bioactivation and then tested for gametocytocidal activity against redox-active PQ metabolites in PQ haemolytic toxicity in
a P. falciparum 3D7A transgenic strain, expressing luciferase G6PDH-deficient individuals3,18, we indirectly assessed the
specifically in the gametocyte stages14, in an improved gameto- contribution of reactive oxygen species to the gametocytocidal
cyte viability assay GC-LUC (gametocyte luciferase)14. As shown activity of huCPR-treated PQ metabolites by using sodium
in Fig. 1g, PQ at 10 μM, as expected, has minimal activity in the pyruvate as a H2O2 scavenger19. As shown in Fig. 1i (compare
absence of any metabolic conversion (red histograms, (−)HLM). huCPR and huCPR + pyruvate), the presence of pyruvate
When gametocytes were treated with OH-PQm, we observed little abrogated OH-PQm gametocytocidal activity (p < 0.01; Mann
to moderate gametocytocidal activity, with only demethoxylated Whitney, Supplementary Table 7). These results establish the
species, 5,6-DPQ and 6OHPQQI, able to reduce gametocyte need for huCPR catalytic activity for OH-PQm mechanism of
viability to less than 50 % of solvent control (Fig. 1g, (−)HLM). action and point to a direct role for huCPR as a redox cycler of
After HLM metabolism (Fig. 1g, blue histograms, (+)HLM), PQ OH-PQm.
N
A
T
U
R
E
C
OM
M
U
N
I
C
A
T
I
ON
S
|
(2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
ww
.n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
re
c
o
m
m
u
n
ic
a
t
i
o
n
s
3

A
R
T
I
C
L
E
N A T U R E C O M MU N ICA T IO N S | h t t p s : / /d o i .o r g / 1 0 .1 0 38 / s 41 46 7 - 01 9 -1 1 2 39 - 0
Redox cycling of PQ metabolites generates gametocytocidal OH-PQm anti-gametocyte activity and that this process occurs
H2O2. To further confirm this observation, we directly measured through generation of H2O2, implying a redox cycling, CPR-
H2O2 generation during drug incubation with human CPR mediated mechanism of parasite death directly attributable to the
(huCPR). To this end, O2 consumption during incubations was generation of H2O2.
recorded using a closed O2 electrode and H2O2 was quantified via
oxygen release after catalase addition. As shown in Fig. 2a, all
metabolites, compared to methanol control, elicited H2O2 pro-
duction, whereas negligible amounts of H2O2 were detected when
PQ was used as a substrate. When sodium pyruvate was added to
CPR reactions, H2O2 production was completely abolished
(Fig. 2b). In addition, we measured H2O2 production in titration
experiments with 5-HPQ as a representative OH-PQm (Supple-
mentary Fig. 5) and observed that as low as 1 nM 5-HPQ is able
to elicit H2O2 production. We confirmed H2O2 involvement in
gametocyte killing by directly assessing its production upon HLM
and CYP2D6 metabolism of PQ and 5-HPQ and 5,6-DPQ as
representative metabolites (Fig. 2c, d and Supplementary Fig. 4
for representative full traces of oxygen content). Collectively,
these data clearly demonstrate that huCPR is required for
H2O2 production by PQ in bone marrow. We reasoned that
spatial co-localisation of parasite and host-dependent metabolic
activating system(s) could account for PQ stage-specificity. For
EEFs this association is obvious with parasite residency within the
hepatocyte; for gametocytes this is less obvious as circulating
gametocytes are rarely in juxtaposition to a potential source of
metabolism. However, P. vivax and P. falciparum gametocytes,
including mature stages20,21, are sequestered/enriched in the bone
marrow22,23, which is also endowed with significant P450-
dependent metabolic capacity24–26. To address this possibility,
the ability of PQ and OH-PQm to generate H2O2 in mouse bone
marrow cell extracts was investigated. Results in Fig. 2e show that
PQ and OH-PQm generated H2O2 when incubated with
a
b
30
CPR
CPR + Pyruvate
30
20
10
0
5,6-DPQ
5-HPQ
20
10
0
PQQI
6OHPQQI
PQ
MeOH
0
1
2
3
4
0
1
2
3
4
Time after catalase addition (min)
Time after catalase addition (min)
c
d
10
Human liver microsomes
CYP2D6 baculosomes
150
100
50
5,6-DPQ
5-HPQ
PQ
8
6
4
2
0
MeOH
0
0
1
2
3
4
0
1
2
3
4
Time after catalase addition (min)
Time after catalase addition (min)
e
f
Mouse bone marrow
Red blood cells
5
4
3
2
1
0
5
5,6-DPQ
5-HPQ
PQ
4
3
2
1
0
MeOH
0
1
2
3
4
0
1
2
3
4
Time after catalase addition (min)
Time after catalase addition (min)
Fig. 2 M e ta b o l i sm - m e d i a t e d g en er a t io n o f H O . H O p r o d u c t i o n w a s m e a s u r e d a s c a t a l a s e - m ed i a t e d o x y ge n r e l e a s e a f t e r c o m p o u n d i n c u ba ti on
2
2
2
2
( 3 0 μM ) w i th a h u m a n C P R , b h u m a n C P R + p y r u v a t e , c h u m a n l i v e r m i c r o s o m e s , d C Y P 2 D6 - e x p re s s i n g b a c u l o s o m e s , e m o u s e b o n e m a r r o w e x t r ac t s o r f
r e d b l o o d c e l l e xt r a c t s. T h e x a x i s w a s a d ju s t e d b y d efin i n g t h e a d d i t i o n o f c a t a l a s e a s t = 0 , a n d t h e c o r r e s p on d i n g y a x i s v a l u e d e fin e d a s 0 n m ol m L−1
T h e a v er a g e o f t w o ( P Q a n d 5 , 6 - DP Q) a n d fiv e ( 5 - HP Q a n d M e OH ) i n d ep e n d e nt m e a s u r e me n t s a r e s h o w n
.
4
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s

N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
A
R
T
I
C
L
E
5,6-DPQ, IC50 0.015 0.004 μM
5-HPQ, IC50 0.115 0.03 μM
PQQI, IC50 0.063 0.016 μM
6OHPQQI, IC50 0.04 0.013 μM
Significantly, our data also shows that OH-PQm exert anti-
malarial activity against liver parasite stages independently of
CYP2D6 (Fig. 2). This finding has significant implications for the
design of improved 8-aminoquinolines which retain antimalarial
activity but which importantly would not be determinately
affected by the CYP2D6 status of the patient.
PQ, IC50 2.59 0.77 μM
100
The described biochemical relay accounts for a unique mode of
action (MoA) involving two host enzymes and has implications
not only for bioactivation but also for pharmacodynamics.
The elucidation of a catalytic (e.g. 1 nM) MoA for OH-PQm
(generating μM levels of H2O2, Supplementary Fig. 5) explains
how single low-dose PQ, down to 0.25 mg/kg, retains transmis-
sion blocking ability in clinical settings30; importantly, this
treatment is also well tolerated in G6PD deficient patients30. The
emphasis on a host-mediated MoA, not directly involving any
parasite targets that are subject to selective pressure, may also
provide an explanation for the apparent lack of resistance to PQ
in the field; in fact, treatment failure so far has only been linked to
CYP2D6 polymorphisms.
50
0
–3
–2
–1
0
1
2
Log[Drug], μM
Fig. 3 D o se - r e sp o n s e v i a b i l i ty o f l a t e s t a ge g a m e t o c yt e s t r e a t e d w i t h
P Q a n d O H - PQ m u p o n r e a c t io n w it h h u m a n C P R . C o m p o u nd s a t
These results do not exclude a role for other human host strong
reducing enzymes in mediating PQ antimalarial activity. How-
ever, we propose that the tissue-specific juxtaposition of CYP2D6
and CPR, in the liver and bone marrow24–26, exquisitely explains
the susceptibility of liver stages, including hypnozoites, of
P. falciparum gametocytes, and of P. vivax gametocytes and
asexual stages, the latter recently shown to be enriched in the
bone marrow23. Moreover, the results presented accurately mir-
ror the clinical efficacy data in that PQ appears to be more active
against gametocytes than liver stages. This observation supports
the validity of the in vitro models used here as proxies for in vivo
conditions.
Overall, these data provide a framework that will make it
possible to test the potential of rationally re-designing PQ ana-
logues that can be activated even against a backdrop of CYP2D6
poor metaboliser status. Furthermore, armed with this knowl-
edge, it will be possible to test if antimalarial activity can ever be
divorced from haemotoxicity.
c
o
n
c
e
n
t
r
a
t
i
o
n
s
r
a
n
g
i
n
g
f
r
o
m
3
0
μ
M
t
o
1
1
.
7
n
M
w
e
r
e
i
n
c
u
b
a
t
e
d
w
i
t
h
h
u
m
a
n
C P R a n d d i l u te d 1 : 3 f o r G C - L U C a s s a y s ( n = 2 , e a c h i n t r ip li c a t e ) . I C
( S . E. M .) w e re c a l c u l a t e d f ro m n o n li n ea r r e gr e s s i on ( c u r v e fit ) i n
G ra ph P a d P r i s m
s
5
0
bone marrow crude cell extracts, whereas no H2O2 could be
detected with equivalent amount of proteins from red blood cells
(to represent the circulating parasite environment) (Fig. 2f).
These results support the view that localised H2O2 generation at
specific anatomical sites is responsible for stage-specific parasite
killing by PQ metabolites.
PQ-metabolite efficacy at pharmacologically-relevant doses.
Next, we determined gametocyte inhibition sensitivity profiles
(IC50) of OH-PQm. Dose-response huCPR reactions were per-
formed prior to GC-LUC assays. As shown in Fig. 3, OH-PQm
are all active at pharmacologically-relevant nanomolar con-
centrations, with 5,6-DPQ being the most active with an IC50 of
15 0.4 nM. Conversely, parental PQ shows an IC50 of 2.59
0.77 μM, in strong agreement with the measured enzyme kinetics.
These results support a catalytic mechanism of action for PQ
metabolites via CPR-mediated redox cycling that is in line with
expected drug metabolite exposure after clinically relevant PQ
Methods
Reagents. All chemical reagents were from Sigma-Aldrich unless otherwise
specified.
Chemical syntheses. Reactions that were air and moisture sensitive were per-
formed under a nitrogen or argon atmosphere. This was achieved with oven dried
or flame dried glassware sealed with a rubber septum. Dry nitrogen gas was
introduced via a manifold or balloon.
doses27–29
.
Reactions were stirred using Teflon-coated magnetic stir bars. Organic solutions
were concentrated using a Büchi rotary evaporator with a diaphragm vacuum
pump. Anhydrous solutions and sensitive liquids were transferred via syringe.
All reagents were purchased from Sigma Aldrich or Alfa Aesar and were used
without purification unless otherwise indicated. 5,6-dimethoxy-8-nitroquinoline
was obtained from WuXi AppTec.
Discussion
Our results unveil a two-step biochemical relay underlying PQ
mode of action relying on CYP2D6 and its redox partner CPR
(Fig. 4). PQ oxidation principally by CYP2D6 generates hydroxyl-
metabolites whose oxidation to quinoneimine generates H2O2.
Quinoneimines in turn are substrates for CPR reducing activity,
thus leading to H2O2 accumulation which can exert anti-parasitic
activity directly through oxidation of protein sulphydryl groups
and damage of Fe- and FeS-containing proteins or through the
generation of superoxide and hydroxyl radicals. The intrinsic
instability of parent hydroxyl-metabolites prevents identification
of all the relevant chemical species in the process. However, it is
important to emphasise that our results show that quinoneimine
metabolites, considered as stable degradation products or markers
for the unstable hydroxylated forms28,29, are also substrates for
redox cycling activities involved in anti-parasitic action. Direct
demonstration of CYP2D6-dependent PQ antimalarial activity
against P. falciparum liver stages supports clinically-led hypoth-
eses for the essentiality of CYP2D6 for PQ activity8.
1H NMR spectra were measured on a Brucker AMX400 (400 MHz) nuclear
magnetic resonance spectrometer. Solvents are indicated in the text. The data for
1H NMR spectra are reported as follows: chemical shifts were described in parts per
million (δ, ppm) downwards from an internal reference of trimethylsilane. 13C
NMR spectra were measured on a Brucker AMX400 (100 MHz) and are reported
in terms of chemical shift (δ, ppm) relative to residual solvent peak. MS and High
resolution mass spectrometry (HRMS) were recorded on a VG analytical 7070E
machine, Frisons TRIO spectrometers or Agilent QTOF 7200 using chemical
ionisation (CI) or electron ionisation (EI). Micromass LCT mass spectrometer used
electron spray ionisation (ESI).
The synthesis of metabolites was based on modifications of literature
procedures (Supplementary Fig. 6)31,32 as described below.
For the synthesis of 5,6-dimethoxyquinolin-8-amine31 5,6-dimethoxy-8-
nitroquinoline (2 g, 8.54 mmol) was dissolved in Tetrahydrofuran (THF, 20 mL). A
solution of sodium hypophosphite (6 g, 68.32 mmol) in 10 mL water was added to
the reaction. The flask was purged with nitrogen. 10% palladium on carbon (0.7 g)
was added to the reaction mixture and purged with nitrogen. The reaction was
allowed to stir vigorously for 10 min. A celite pad was prepared and the reaction
mixture was filtered through the pad of celite and washed through with 20 mL
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
ON
S
|
(
2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
5

A
R
T
I
C
L
E
N A T U R E C O M MU N ICA T IO N S | h t t p s : / /d o i .o r g / 1 0 .1 0 38 / s 41 46 7 - 01 9 -1 1 2 39 - 0
Primaquine mode of action
STEP 1
Cytochrome
P450 reductase
NADP+
Cytochrome
P450 2D6
FAD
O2
e–
NADPH
PQ
FMN
OH-PQm
HEME
e–
STEP 2
Cytochrome
P450 reductase
H2O2
NADP+
e–
antiparasitic activity
O2
OH-PQm
O=PQm
FAD
e–
FMN
NADPH
H2O2
H2O2
e–
Liver stages
Gametocytes
Fig. 4 S c h e ma t i c r e p r e se n t a ti o n o f p r i m a q u i n e m o d e o f a c t i o n . T h e r e s u l t s p r es e n t e d i n t h i s w o r k s u p p o r t a t w o - s t e p b i o ch e m i c a l r e l a y m e c h a n ism f or P Q
m o d e o f a c t i o n . I n S te p 1 , P Q i s c o n v e rt e d i n t o h y d r o x y l at e d m e t a b o l i t es ( O H - P Qm ) t h r o u gh t h e C P R / C YP 2 D6 m e t a b o l i c c o m p l ex . I n S t e p 2 , m e t a b ol i te s
t he n u n d e r g o s p o n t a n e o u s o x i d a ti o n t o q u i n o n e im i ne s ( O = P Q m ) w i t h c o n c o m i ta n t g e n e r a t i o n o f H O . H u m a n C P R t h e n r e c e i v e s t w o e l e c t r on s f r o m
2
2
N A D P H a n d o r c h es t r a te s o n e - e le c t r o n t r a n s f e r s v i a F A D/ F MN c o f a c t o r s t o q u i n o n e i n i me s , t h u s r e d u ci n g t h e m b a c k t o t h e h y d r o x y l f o r m s a nd
p e r p e t u a t in g a c a t a l y t i c c y c l e w h ic h b r i n g s a b o u t H O a c c u m ul a t io n a t s i t es o f m e t a b o l i c t r a n s f o r m a t i o n ( l i v e r , b o n e m a r r o w a n d p o s s i b l y o t he r s) .
2
2
Plasmodium p a r a s i te s p r e s e n t a t t h e s e l o c a t i o n s a r e t h e n k i l l ed b y H O a c t i o n
2
2
chloroform. Sodium hydroxide (2 M, 50 mL) was added to the filtrate. The filtrate
was extracted with chloroform (3 × 100 mL) and the combined organic extracts
were washed with water (2 × 50 mL), dried over magnesium sulphate and filtered.
The chloroform was removed under reduced pressure to obtain the pure product
(1.72 g, 98.5%) as a red-orange solid; Rf = 0.15, 30 % ethyl acetate in hexane; 1H
NMR (400 MHz, CDCl3) δ 8.62 (dd, J = 4.1, 1.6 Hz, 1 H), 8.33 (dd, J = 8.5, 1.6 Hz,
1 H), 7.35 (dd, J = 8.5, 4.1 Hz, 1 H), 6.73 (s, 1 H), 4.87 (s, 2 H), 3.96 (s, 3 H) and
3.89 (s, 3 H); 13C NMR (100 MHz, CDCl3) δ 149.6, 145.6, 141.5, 133.8, 133.2, 129.9,
124.5, 121.6, 99.1, 61.6 and 56.8; IR νmax (neat)/cm−1 3443, 3337 (NH), 3192 (CH)
and 1623 (C=C); HRMS calculated for C11H12N2O2 [M + H]+ 205.0972 found
205.0977.
For the synthesis of 4-bromo-1-phthalimidopentane, potassium phthalimide
(10 g, 53.99 mmol) was dissolved in acetone. 1,4-Dibromopentane (9.57 mL,
70.19 mmol) was added to the mixture which was heated to reflux for 24 h. The
reaction was cooled and filtered. The acetone was removed under reduced pressure
and the crude product was purified via flash chromatography resulting in (14.62 g
91%) a clear light yellow oil. Rf = 0.43, 20% ethyl acetate in hexane; 1H NMR (400
MHz, CDCl3) δ 7.88–7.81 (m, 2 H), 7.76–7.70 (m, 2 H), 4.22–4.10 (m, 1 H), 3.72
(dd, J = 8.5, 4.7 Hz, 2 H), 2.02–1.74 (m, 4 H) and 1.70 (d, J = 6.7 Hz, 3 H); 13C
NMR (100 MHz, CDCl3) δ 168.4, 134.1, 132.1, 123.3, 50.6, 38.1, 37.2, 27.1 and
26.5; HRMS Calculated for C13H14BrNO2 [M + H]+ 296.0281 found 296.0280.
5,6-Dimethoxy-8-aminoquinoline (500 mg, 2.45 mmol) was added to a dry 50 mL
two neck flask along with a dry magnetic bar. A dry condenser and equilibrating
dropping funnel were attached to the flask and the second neck was sealed with a
rubber septum. The system was purged with dry argon and sealed with a balloon
attached to the dropping funnel. 1-Phthalimido-4-bromopentane (942.6 mg, 3.18
mmol) was added directly to the reaction via syringe and the reaction was then
heated to 150 °C to produce a dark paste. Triethylamine (444 µL, 3.18 mmol)
was added to the reaction via the dropping funnel over a 1.5 h period. The reaction
was left to stir for a further 1.5 h. Additional 1-phthalimido-4-bromopentane
(1.23 g, 4.16 mmol) was added directly to the reaction via syringe. Triethylamine
(290.3 µL, 2.08 mmol) was added to the reaction via the dropping funnel over
30 min. The reaction was left to stir for a further 2 h. 1-Phthalimido-4-
bromopentane (235.7 mg, 795.7 µmol) was added to the reaction directly via
syringe. Triethylamine (122.9 µL, 881.4 µmol) was added to the reaction via the
dropping funnel over 30 min. The reaction was left to stir for a further 2 h and
monitored by thin layer chromatography (TLC). The reaction was cooled, diluted
with acetone and filtered. The acetone was removed, resulting in a dark oil which
was dissolved in chloroform (100 mL). The organic layer was washed with water
(3 × 50 mL), dried over magnesium sulphate and filtered. The crude product was
pre-absorbed onto silica and purified via flash chromatography to obtain the
product (790.5 mg, 77%) as a yellow oil; Rf = 0.3, 30% Ethyl Acetate in Hexane; 1H
NMR (400 MHz, CDCl3) δ 8.53 (dd, J = 4.1, 1.6 Hz, 1 H), 8.27 (dd, J = 8.5, 1.6 Hz,
1 H), 7.84–7.78 (m, 2 H), 7.72–7.66 (m, 2 H), 7.33 (dd, J = 8.5, 4.1 Hz, 1 H), 6.41 (s,
1 H), 5.89 (d, J = 6.4 Hz, 1 H), 3.99 (s, 3 H), 3.86 (s, 3 H), 3.74 (t, J = 7.1 Hz, 2 H),
3.71–3.63 (m, 1 H), 1.99–1.62 (m, 4 H) and 1.30 (d, J = 6.3 Hz, 3 H); 13C NMR
(100 MHz, CDCl3) δ 168.6, 150.0, 144.8, 141.9, 134.0, 133.7, 132.2, 131.3, 129.8,
124.6, 123.3, 121.7, 94.3, 61.6, 57.0, 48.1, 38.1, 34.1, 25.5 and 20.8; IR νmax (neat)/
cm−1 3390 (NH), 2964, 2936 (CH) and 1707 (C=O); HRMS Calculated for
C
24H25N3O4 [M + H]+ 420.1923 found 420.1922.
6
N
A
T
U
R
E
C
O
M
M
UN
I
C
A
T
I
O
N
S
|
(2 0 1 9 ) 1 0 : 3 2 2 6 | h t t p s : / / d o i . o r g / 1 0 . 1 0 3 8 / s 41 4 6 7 - 0 1 9 - 1 1 2 3 9 - 0 | w ww .n a t u r e . c o m / n a t u re c o m m u n i c a t i o n s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 2 3 9 - 0
A R T I C L E
For the synthesis of 5,6 dimethoxy primaquine31, 5,6 dimethoxy-phthaloyl
primaquine (790.5 mg, 1.88 mmol) was dissolved in 25 ml of ethanol in a 50 mL
round bottom, a condenser was attached and the system was purged with nitrogen.
Sixty-five percent hydrazine monohydrate (465.5 µL, 6.22 mmol) was added to the
reaction and refluxed (at 100 °C) for 6 h. A solid precipitate was observed. The
reaction was cooled and filtered. The ethanol was removed and 30% potassium
hydroxide (100 mL) was added to the residue. The mixture was extracted with
diethyl ether (3 × 50 mL). The combined organic layers were washed with water
(50 mL), dried over magnesium sulphate and filtered. A solution of 89%
phosphoric acid (187 µL, 2.73 mmol) in ethanol (5 mL) was added drop wise to the
ether solution. A red/orange precipitate was observed. The solvent was removed
under vacuo and the precipitate was recrystallised with ethanol to obtain an
orange/red solid (501 mg, 92%); 1H NMR (400 MHz, DMSO) δ 8.57 (dd, J = 4.1,
1.5 Hz, 1 H), 8.23 (dd, J = 8.5, 1.5 Hz, 1 H), 7.47 (dd, J = 8.5, 4.1 Hz, 1 H), 6.55 (s,
1 H), 6.02 (s, 2 H), 3.94 (s, 3 H), 3.75 (s, 3 H), 3.80–3.71 (m, 1 H), 2.80 (t, J = 6.3
Hz, 2 H), 1.79 –1.55 (m, 4 H) and 1.23 (d, J = 6.2 Hz, 3 H); 13C NMR (100 MHz,
DMSO) δ 149.8, 144.6, 141.5, 132.6, 130.0, 129.2, 123.8, 122.0, 94.0, 60.9, 56.5, 46.9,
38.9, 33.1, 24.1 and 20.4; HRMS Calculated for C24H25N3O4 [M + H]+ 420.1923
found 420.1922.
Two general procedures were used for demethylation, as indicated. The general
procedure 1 was based on a modified procedure of Allahyari et al.32. Ten milligram
of 5,6 dimethoxy primaquine was dissolved in 1 mL of 48% hydrogen bromide
solution (ca 6 M) and sealed under argon in a sealed tube. The reaction was heated
to 120 °C on a pre-heated mantle for 20 mins. Under a flow of nitrogen, 5 mL of
water was added and the product(s) were purified via HPLC. For the general
procedure 2, 10 mg of 5,6 dimethoxy primaquine was dissolved in 1 mL of 48%
hydrogen bromide solution and the solution was allowed to stir for 6 h. Excess
reagent was removed under vacuum to give a brown solid that was stored under
nitrogen.
compounds were pre-incubated with paroxetine (10 μM) for 15 min at 37 °C before
addition of compounds and further incubation 37 °C for 2 h. Recombinant
huCPR17 was used for in vitro redox cycling of compounds. Reactions were per-
formed as described above with minor modifications. A master mix without
baculosomes was prepared as above; huCPR were added in 10 μl of reaction buffer
(200 nM final concentration) and reactions started by adding 40 μl of compound/
NADP+ mix. For H2O2 scavenging, sodium pyruvate (10 mM) was used, without
pre-incubation. All experiments contained control reactions with solvent (MeOH:
water 50:50) only. After incubations, reaction mixes were spun down, supernatants
collected, diluted 2:3 and seeded into 96-well plates for GC-LUC assay.
Parasite culture, drug treatments and gametocyte luciferase assay (GC-LUC).
A P. falciparum 3D7A33 transgenic derivative 3D7elo1-pfs16-CBG99 was used14,34
specifically expressing the CBG99 luciferase in gametocytes. Parasites were cul-
,
tured35 in human 0 + erythrocytes at 5% haematocrit under 5% CO2, 2% O2, 93%
N2. Cultures were grown in complete medium (CM) containing RPMI 1640
medium (Gibco) supplemented with 25 mM Hepes (VWR), 50 μg mL−1 hypox-
anthine, 0.25 mM NaHCO3, 50 μg mL−1 gentamicin sulfate, and 10% pooled heat
inactivated AB + human serum. Gametocyte viability was evaluated by the GC-
LUC assay14,34. Parasites were quickly harvested in 2× CM before addition to
96-well plates containing control, CYP2D6- or huCPR-treated compounds (10 μM
parental compound final concentration) in aqueous solution and incubatied at
37 °C for 72 h. Drug-treated gametocytes were then transferred to 96-well white
plate; D-Luciferin, 1 mM in 0.1 M citrate buffer pH 5.5 (Promega), was added in a
1:1 ratio and luminescence measurements were recorded after 10 min on a
FLUOstar Omega plate reader (BMG Labtech). Viability was expressed as % via-
bility as compared to solvent treated controls.
HPLC conditions were as follows: a Phenomenex Jupiter Proteo 90 A column,
250 × 10 mm, 10 Micron was used for purification using a gradient system based on
the following conditions; Initial (time 0) solvent mix 5% acetonitrile, 95% 0.05%
trifluoracetic acid (TFA) in water; 20 mins, 25% acetonitrile, 75% 0.05% TFA in water;
Liver stages assay. Micropatterned hepatocyte-fibroblast co-cultures were
established as previously described12,36. Briefly, soft lithography techniques were
used to pattern rat tail type I collagen (Corning) into 500 µm diameter islands on
the surface of glass bottom 96-well plates. Cryopreserved primary human hepa-
tocytes (Bioreclamation IVT) were thawed and pelleted through centrifugation at
100 g for 6 min, assessed for viability using trypan blue exclusion (70–90% viabi-
lity) and 10,000 hepatocytes were seeded onto the collagen islands in serum-free
DMEM (Dulbecco’s Modified Eagle’s medium) with 1% Penstrep. Two to three
hours later, cells were washed with DMEM containing 1% Penstrep, and media was
replaced with hepatocyte culture media. The following day, each well was infected
with 70,000 fresh P. falciparum sporozoites. Three hours later, cultures were
washed with DMEM containing 3% Penstrep and 0.1% Fungizone, and 7000 3T3
J2 mouse fibroblasts were added to establish the co-culture. Drug was administered
during daily media changes for 3 days. The impact on hepatocyte infection was
measured by enumeration of exoerythrocytic forms on day 3.5 post infection,
through staining for PfHSP70 and visualisation with a Nikon Eclipse Ti fluorescent
microscope.
20.10 min 5% acetonitrile, 95% 0.05% TFA in water; flow rate was 5 ml min−1
.
All samples were dried by blowing off the solvent under a flow of nitrogen 5,6-
dihydroxyprimaquine (5,6-DPQ) was prepared according to general procedure 2;
this compound was very unstable rapidly oxidising to 5-hydroxy quinoneimine
(5-HPQ) in solutions exposed to air. (MS for C14H19N3O2 [M + H] + found
262.33). The metabolite was stored under nitrogen in a sealed tube.
For the synthesis of 5-hydroxy primaquine (5-HPQ) general procedure 1 along
with HPLC purification and drying was used (see HPLC conditions above).
5-Hydroxy 6-Methoxy Primaquine = 10.06 min. The 5-hydroxy 6-methoxy
primaquine was very unstable as it readily oxidises to the quinoneimine form.
LCMS has shown that this is present within the reaction mixture, however, upon
isolation and re-evaluation using the HPLC conditions, the retention time at 10.06
min corresponding to the product is no longer observed. Once the quinoneimine
has formed from 5-OH primaquine, it is readily converted to the 6-hydroxy form
(6OHPQQI; Rt = 12.55 mins) by a demethylation reaction (through reaction with
water).
For the synthesis of Primaquine quinone-imine (PQQI), general procedure 1
provides the -hydroxy primaquine which can be allowed to oxidise to the
quinoneimine in aqueous solution. HPLC purification and drying (see above) were
used for isolation. The quinoneimine was purified with two HPLC purification
runs. Retention time: = 7.76 min 1H NMR (500 MHz, D2O) δ 9.03 (d, J = 3.6 Hz,
1 H), 8.59 (d, J = 6.6 Hz, 1 H), 7.96 (dd, J = 8.0, 4.8 Hz, 1 H), 6.99 (s, 1 H), 4.22 (s,
3 H), 3.07 (t, J = 7.6 Hz, 3 H), 2.11–1.95 (m, 3 H), 1.88–1.72 (m, 3 H) and 1.58 (d,
J = 6.5 Hz, 3 H); MS for C15H19N3O2 [M + H]+ found 274.38.
For the synthesis of 6-hydroxy primaquine quinone imine (6OHPQQI), general
procedure 2 along with HPLC purification and drying (see section 1.1.3) was used.
This compound could also be produced by reaction of the 6-methoxy
quinoneimine in aqueous solution followed by HPLC purification. (See analysis
below) Retention time: = 12.55 min. 1H NMR (500 MHz, D2O) δ 8.86 (dd, J = 4.9,
1.7 Hz, 1 H), 8.40 (dd, J = 7.9, 1.7 Hz, 1 H), 7.75 (dd, J = 7.9, 4.9 Hz, 1 H), 6.12 (s, 1
H), 4.11 (dd, J = 13.1, 6.7 Hz, 1 H), 3.05 (t, J = 7.4 Hz, 2 H), 1.92–1.74 (m, 4 H) and
1.42 (d, J = 6.5 Hz, 3 H); MS for C14H17N3O2 [M + H]+ found 260.35.
Determination of oxygen consumption and H2O2 production. Oxygen con-
sumption and H2O2 production measurements were performed using the Oxy-
therm system and O2 View software package v.2.06 (Hansatech Instruments Ltd).
Compounds ability to generate hydrogen peroxide after in vitro metabolism or
huCPR reaction, in the presence of regeneration system or 100 μM NADPH as
indicated, was assessed indirectly by measuring catalase-mediated oxygen release.
Once the kinetic trace for oxygen concentration within the mixture had reached a
plateau for at least 3 min, catalase (from bovine liver, prepared in 50 mM potas-
sium phosphate buffer, pH 7.0; final assay concentration 10 µg ml−1) was added
and data recorded for a further 6 min. Oxygen concentration was recorded as nmol
mL−1. To allow for easier comparison of individual traces, the x axis was adjusted
by defining the addition of catalase as t = 0, and the corresponding y axis value
defined as 0 nmol mL−1
.
For bone marrow experiments, femurs were dissected from mice following
schedule 1 procedure. The schedule 1 procedure was undertaken with local (LSTM
and UoL Animal Welfare Ethics Review Boards) and national (Home Office
licence) authorisation. Bone marrow cells were flushed out with cold Ringer’s
solution pH 7.4 (125 mM NaCl, 1.5 mM CaCl2, 5 mM KCl, 0.8 mM Na2HPO4)
using 2 ml syringe connect with a 25GA needle. The bone marrow cells were
washed twice and resuspended with cold Ringer’s solution. Cells were lysed by
sonication and the protein concentration of crude extracts measured. Oxygen
measurements were performed at 37 ˚C in 0.4 ml samples containing Ringer’s
solution pH 7.4, 8.5 mg ml−1 bone marrow extracts, 1X regeneration system
(Thermo Scientific), 30 µM PQ metabolites, 30 µM NADP+. The assay mixture
without NADP+ was pre-incubated in the Oxytherm’s chamber at 37˚C while
recording oxygen content. Then, the reaction was initiated by addition of NADP+.
After 30, 10 µl of 5 mg ml−1 catalase (prepared in 50 mM potassium phosphate
buffer pH 7.0) was added to release O2 from H2O2. Experiments with red blood
cells were performed as above with same amount of protein extracts.
In vitro metabolism and redox cycling reactions. Human liver microsomes
(HLM, BD Biosciences) and CYP2D6-expressing baculosomes from the Vivid
CYP450 kit (Life Technologies) were used for compound metabolic conversions as
per manufacturer’s instructions with minor modifications. For HLM, compounds
(30 μM final concentration) were incubated in the presence of NAPDH Regen-
eration System Solution A and B (BD Biosciences) at 37 °C for 2 h in a total volume
of 100 μl of phosphate reaction buffer. For CYP2D6 baculosomes reaction, a 2×
master mix (50 μl) was prepared containing reaction buffer, baculosomes, regen-
eration system and NADP+ (all from Life technologies). Compounds (30 μM final
concentration) were prepared in 40 μl reaction buffer containing NADP+ as per kit
protocol. The total volume was brought to 100 μl with reaction buffer and reactions
incubated at 37 °C for 2 h. For CYP2D6 inhibition, we first determined paroxetine
inhibitory profile using Vivid CYP2D6 Blue kit (Life Technologies) according to
manufacturer’s instructions. For gametocytocidal assays, master mixes without
Steady-state kinetic measurements. Kinetic measurements were determined
following Tsukamoto et al.37 and the assay conditions were optimised according to
Döhr et al.38 with minor modifications. Briefly, huCPR kinetics of interaction with
N
A
T
U
R
E
C
OM
M
U
N
I
C
A
T
I
O
N
S
|
(2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
ww
.n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
7

A
R
T
I
C
L
E
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
PQ and OH-PQm were measured as NADPH consumption via recording of
change in absorbance at 340 nm on a microplate spectrophotometer (Thermo
Electron Varioskan) at 25 °C. The activity assay of 0.3 ml contained 0.3 M potas-
sium phosphate (pH 7.7), 25 nM huCPR, 100 µM NADPH, and various con-
centrations of primaquine derivatives (0–200 µM 5-HPQ, 0–100 µM 5,6-DPQ,
0–150 µM PQQI, 0–150 µM 6OHPQQI, or 0–1000 µM PQ). Reactions were initi-
ated by adding NADPH to a final concentration of 100 µM after equilibrating the
assay mixture with all other components at 25 °C for 2 min. Data were recorded
every 5 s over 4 min. The Michaelis–Menten equation was used to determine Km
and kcat values.
17. Gutierrez, A. et al. Trp-676 facilitates nicotinamide coenzyme exchange in the
reductive half-reaction of human cytochrome P450 reductase: properties of
the soluble W676H and W676A mutant reductases. Biochemistry 39,
15990–15999 (2000).
18. Vásquez-Vivar, J. & Augusto, O. Oxidative activity of primaquine metabolites on
rat erythrocytes in vitro and in vivo. Biochem. Pharmacol. 47, 309–316 (1994).
19. Giandomenico, A. R., Cerniglia, G. E., Biaglow, J. E., Stevens, C. W. & Koch,
C. J. The importance of sodium pyruvate in assessing damage produced by
hydrogen peroxide. Free Radic. Biol. Med. 23, 426–434 (1997).
20. Aguilar, R. et al. Molecular evidence for the localization of Plasmodium
falciparum immature gametocytes in bone marrow. Blood 123, 959 (2014).
21. De Niz, M. et al. Plasmodium gametocytes display homing and vascular
transmigration in the host bone marrow. Sci. Adv. 4, eaat3775 (2018).
22. Joice, R. et al. Plasmodium falciparum transmission stages accumulate in the
human bone marrow. Sci. Transl. Med. 6, 244re245–244re245 (2014).
23. Baro, B. et al. Plasmodium vivax gametocytes in the bone marrow of an acute
malaria patient and changes in the erythroid miRNA profile. PLoS Negl. Trop.
Dis. 11, e0005365 (2017).
Statistics. Statistical analyses were done using GraphPad Prism version
5.04 and version 7 software (GraphPad Software, San Diego, California, USA,
www.graphpad.com). Significance was calculated by two-tailed Mann Whitney
test. One-way or two-way ANOVA was used as appropriate. Extended statistics
tables for each analysis are provided as Additional Information.
Reporting summary. Further information on research design is available in
24. Ganousis, L. G., Goon, D., Zyglewska, T., Wu, K. K. & Ross, D. Cell-specific
metabolism in mouse bone marrow stroma: studies of activation and
detoxification of benzene metabolites. Mol. Pharmacol. 42, 1118 (1992).
25. Alonso, S. et al. Human bone marrow niche chemoprotection mediated by
cytochrome p450 enzymes. Oncotarget 6, 14905–14912 (2015).
26. Zhang, Y. et al. Suppression of cytochrome P450 reductase enhances long-
term hematopoietic stem cell repopulation efficiency in mice. PLoS ONE 8,
e69913 (2013).
the Nature Research Reporting Summary linked to this article.
Data availability
The source data underlying Figs. 1a, 2a–d, 6d, h and 7c and Supplementary Figs. 1a and
5d are provided as a Source Data file.
Received: 14 September 2018 Accepted: 2 July 2019
27. Ward, S. A. et al. Pharmacokinetics of primaquine in man. II. Comparison of
acute vs chronic dosage in Thai subjects. Br. J. Clin. Pharm. 19, 751–755
(1985).
28. Potter, B. M. et al. Differential CYP 2D6 metabolism alters primaquine
pharmacokinetics. Antimicrob. Agents Chemother. 59, 2380–2387 (2015).
29. Fasinu, P. S. et al. Enantioselective metabolism of primaquine by human
CYP2D6. Malar. J. 13, 507 (2014).
30. Bancone, G. et al. Single low dose primaquine (0.25mg/kg) does not cause
clinically significant haemolysis in G6PD deficient subjects. PloS ONE 11,
e0151898 (2016).
31. Elderfield, R. C., Mertel, H. E., Mitch, R. T., Wempen, I. M. & Werble, E.
Synthesis of primaquine and certain of its analogs. J. Am. Chem. Soc. 77,
4816–4819 (1955).
32. Allahyari, R., Strother, A., Fraser, I. M. & Verbiscar, A. J. Synthesis of certain
hydroxy analogs of the antimalarial drug primaquine and their in vitro
methemoglobin-producing and glutathione-depleting activity in human
erythrocytes. J. Med. Chem. 27, 407–410 (1984).
References
1. Ashley, E. A., Recht, J. & White, N. J. Primaquine: the risks and the benefits.
Malar. J. 13, 418 (2014).
2. Marcsisin, S. R., Reichard, G. & Pybus, B. S. Primaquine pharmacology in the
context of CYP 2D6 pharmacogenomics: current state of the art. Pharm. Ther.
161, 1–10 (2016).
3. Tarlov, A. R., Brewer, G. J., Carson, P. E. & Alving, A. S. Primaquine
sensitivity: glucose-6-phosphate dehydrogenase deficiency: an inborn error of
metabolism of medical and biological significance. Arch. Intern. Med. 109,
209–234 (1962).
4. Strother, A., Fraser, I. M., Allahyari, R. & Tilton, B. E. Metabolism of 8-
aminoquinoline antimalarial agents. Bull. World Health Organ. 59, 413–425
(1981).
33. Walliker, D. et al. Genetic analysis of the human malaria parasite Plasmodium
falciparum. Science 236, 1661 (1987).
5. Vásquez-Vivar, J. & Augusto, O. Hydroxylated metabolites of the antimalarial
drug primaquine. Oxidation and redox cycling. J. Biol. Chem. 267, 6848–6854
(1992).
6. Vale, N., Moreira, R. & Gomes, P. Primaquine revisited six decades after its
discovery. Eur. J. Med. Chem. 44, 937–953 (2009).
34. D’Alessandro, S. et al. A chemical susceptibility profile of the Plasmodium
falciparum transmission stages by complementary cell-based gametocyte
assays. J. Antimicrob. Chemother. 71, 1148–1158 (2016).
35. Trager, W. & Jensen, J. B. Human malaria parasites in continuous culture.
Science 193, https://doi.org/10.1126/science.781840 (1976).
7. Raphemot, R., Posfai, D. & Derbyshire, E. R. Current therapies and future
possibilities for drug development against liver-stage malaria. J. Clin. Investig.
126, 2013–2020 (2016).
36. March, S. et al. Micropatterned coculture of primary human hepatocytes and
supportive cells for the study of hepatotropic pathogens. Nat. Protoc. 10, 2027
(2015).
8. Bennett, J. W. et al. Primaquine failure and cytochrome P-450 2D6 in
Plasmodium vivax malaria. New Engl. J. Med. 369, 1381–1382 (2013).
9. Ganesan, S., Tekwani, B. L., Sahu, R., Tripathi, L. M. & Walker, L. A.
Cytochrome P(450)-dependent toxic effects of primaquine on human
erythrocytes. Toxicol. Appl. Pharmacol. 241, 14–22 (2009).
10. Pybus, B. S. et al. CYP450 phenotyping and accurate mass identification of
metabolites of the 8-aminoquinoline, anti-malarial drug primaquine. Malar. J.
11, 259 (2012).
37. Tsukamoto, M., Tampo, Y. & Yonaha, M. Lucigenin reduction by NADPH-
cytochrome P450 reductase and the effect of phospholipids and albumin on
chemiluminescence. Biochem. Mol. Biol. Int. 45, 115–123 (1998).
38. Döhr, O., Paine, M. J. I., Friedberg, T., Roberts, G. C. K. & Wolf, C. R.
Engineering of a functional human NADH-dependent cytochrome
P450 system. Proc. Natl Acad. Sci. USA 98, 81 (2001).
11. Committee, W. H. O. M. P. A. & Secretariat Malaria Policy Advisory
Committee to the WHO: conclusions and recommendations of September
2012 meeting. Malar. J. 11, 424 (2012).
12. March, S. et al. A microscale human liver platform that supports the hepatic
stages of Plasmodium falciparum and vivax. Cell Host Microbe 14, 104–115
(2013).
13. Llerena, A., Dorado, P. & Peñas-Lledó, E. M. Pharmacogenetics of
debrisoquine and its use as a marker for CYP2D6 hydroxylation capacity.
Pharmacogenomics 10, 17–28 (2009).
14. Cevenini, L. et al. Multicolor bioluminescence boosts malaria research:
quantitative dual-color assay and single-cell imaging in plasmodium
falciparum parasites. Anal. Chem. 86, 8814–8821 (2014).
15. Bertelsen, K. M., Venkatakrishnan, K., von Moltke, L. L., Obach, R. S. &
Greenblatt, D. J. Apparent mechanism-based inhibition of human CYP2D6
in vitro by paroxetine: comparison with fluoxetine and quinidine. Drug Metab.
Dispos. 31, 289 (2003).
Acknowledgements
We thank the staff and patients of Ward 7Y and the Gastroenterology Unit, Royal
Liverpool Hospital, for their generous donation of blood. We thank Mrs Jill Davies and
Mrs Alison Ardrey for preparing mice for bone marrow experiments; Dr Sitthivut
Charoensutthivarakal and Dr Weiqian David Hong for useful discussion on redox
chemistry and advice on compound handling; Dr Eva Caamano and Dr Ghaith
Aljayyoussi for help with statistical analyses. G.A.B., S.A.W., P.O.N., M.J.I.P. and D.A.B.
wish to acknowledge support from the UK Medical Research Council (MC_PC_16052,
MC_PC_14111 and MR/L000644/1). D.A.B. is grateful to the Wellcome Trust for sup-
port (106240/Z/14/Z).
Author contributions
G.C. designed and performed experiments, analysed data and wrote the paper with
contributions from all authors. P.J., R.S.P., A.S., S.M. and A.B.M. designed and performed
experiments, and analysed data. M.H.L.W. and S.L. synthesised primaquine metabolites.
M.J.I.P. produced recombinant human CPR and gave conceptual advice on enzyme
biochemistry. P.A. and G.C. produced the transgenic parasite strain used in this study.
16. Pandey, A. V. & Sproll, P. Pharmacogenomics of human P450 oxidoreductase.
Front. Pharmacol. 5, 103 (2014).
8
N
A
T
UR
E
C
O
M
M
UN
I
C
A
TI
O
N
S
|
(2 0 1 9 ) 1 0 : 3 2 2 6 | h t t p s : / / d o i . o r g / 1 0 . 1 0 3 8 / s 41 4 6 7 - 0 1 9 - 1 1 2 3 9 - 0 | w ww .n a t u r e . c o m / n a t u re c o m m u n i c a t i o n s

N A T UR E C O M M UN I CA T I O NS | h t t p s : //d o i . o r g / 10 . 1 0 38 / s 4 1 4 6 7 - 01 9 - 1 1 2 3 9 - 0
A R T I C L E
D.A.B., S.B., P.M.O., S.A.W. and G.A.B. provided guidance and gave conceptual advice.
G.A.B conceived the study.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Open Access This article is licensed under
a Creative Commons
Additional information
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467-
019-11239-0.
Competing interests: S.N.B. is a co-founder of Ascendance, which commercially
manufactures and distributes micropatterned co-cultures. All other authors declare no
competing interests.
Reprints and permission information is available online at http://npg.nature.com/
reprintsandpermissions/
Peer review information: Nature Communications thanks Katja Becker and Larry
Walker for their contribution to the peer review of this work. Peer reviewer reports are
available.
© The Author(s) 2019
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
1
9
)
1
0
:
3
2
2
6
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
9
-
1
1
2
3
9
-
0
|
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
9