Novel roflumilast analogs as soft PDE4 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2014.07.016

PDE4 inhibitors are of high interest for treatment of a wide range of inflammatory or autoimmune diseases. Their potential however has not yet been realized due to target-associated side effects, resulting in a low therapeutic window. We herein report the design, synthesis and evaluation of novel PDE4 inhibitors containing a γ-lactone structure. Such molecules are designed to undergo metabolic inactivation when entering circulation, thereby limiting systemic exposure and reducing the risk for side effects. The resulting inhibitors were highly active on both PDE4B1 and PDE4D2 and underwent rapid degradation in human plasma by paraoxonase 1. In contrast, their metabolites displayed markedly reduced permeability and/or on-target activity.

Full text of DOI:10.1016/j.bmcl.2014.07.016

Bioorganic & Medicinal Chemistry Letters 24 (2014) 4594–4597
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel Roflumilast analogs as soft PDE4 inhibitors
,
⇑
Sandro Boland , Jo Alen , Arnaud Bourin, Karolien Castermans, Nicki Boumans, Laura Panitti,
Jessica Vanormelingen, Dirk Leysen, Olivier Defert
Amakem N.V. Agoralaan A bis, 3590 Diepenbeek, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
PDE4 inhibitors are of high interest for treatment of a wide range of inflammatory or autoimmune dis-
eases. Their potential however has not yet been realized due to target-associated side effects, resulting
in a low therapeutic window. We herein report the design, synthesis and evaluation of novel PDE4 inhib-
Received 16 May 2014
Revised 5 July 2014
Accepted 7 July 2014
Available online 12 July 2014
itors containing a c-lactone structure. Such molecules are designed to undergo metabolic inactivation
when entering circulation, thereby limiting systemic exposure and reducing the risk for side effects.
The resulting inhibitors were highly active on both PDE4B1 and PDE4D2 and underwent rapid degrada-
tion in human plasma by paraoxonase 1. In contrast, their metabolites displayed markedly reduced per-
meability and/or on-target activity.
Keywords:
PDE4 inhibitors
Soft drug
SAR
Ó 2014 Elsevier Ltd. All rights reserved.
Esterase
COPD
Cyclic nucleotides, including cyclic adenosine 30,50-monophos-
phate (cAMP) and cyclic guanosine 30,50-monophosphate (cGMP),
play a key role in mediating cellular responses to various hormones
and neurotransmitters. The levels of cAMP and cGMP are tightly
regulated in their synthesis and degradation.1 Cyclic nucleotide
phosphodiesterases (PDE) represent a superfamily of intracellular
enzymes that degrade cyclic nucleotides.2 In humans, 21 PDE
genes are known and can be classified into 11 families, which differ
in their intracellular localization and affinity for cAMP and cGMP.
The cAMP-specific phosphodiesterase 4 (PDE4) family, encoded
by four genes (PDE4A-D), contains over 20 identified isoforms
and is the largest of the 11 PDE families.1 PDE4 isoforms are widely
distributed and highly present among pro-inflammatory, inflam-
matory and immune cells, where they represent the main
cAMP-metabolizing enzymes.1,3 As increased cAMP levels inhibit
inflammatory cell function, PDE4 inhibitors are of potential inter-
est for the treatment of a wide range of inflammatory or autoim-
mune diseases,1 including chronic obstructive pulmonary disease
(COPD),4 asthma,3,4 dry eye disease,5 psoriasis6 or inflammatory
bowel disease (IBD).7
necrosis, has been reported in animal models.8,9 Such side effects
can at least in part be associated with on-target activity. For
instance, a correlation has been found between the emetic effect
of PDE4 inhibitors and their occupancy of the high affinity Rolipram
(Fig. 1) binding site of PDE4 in brain.10 In spite of such issues, the
PDE4 inhibitor Roflumilast (Fig. 1) has been approved in the EU
(as DaxasÒ) and in the US (as DalirespÒ) for treatment of severe
COPD associated with chronic bronchitis and a history of exacerba-
tions.11 Apremilast (OtezlaÒ) has been recently approved for the
treatment of psoriatic arthritis.12 Many development programs of
PDE4 inhibitors have however been discontinued due to side
effects,8,9 and the full potential of PDE4 inhibitors is yet to be
exploited.
Several attempts have been made to mitigate the side effects
resulting from PDE4 inhibition. It has been suggested that occu-
pancy of the high affinity binding site for Rolipram was responsible
of emetic effects and that PDE4 inhibitors targeting the low affinity
binding site of Rolipram would display an improved therapeutic
window.10 However, Cilomilast, which displays selectivity towards
the low affinity binding site still causes side effects.13 Isoform
Safety-related issues represent a significant challenge in the
development of PDE4 inhibitors. Reported side effects resulting
from systemic exposure to PDE4 inhibitors include emesis, nausea,
dyspepsia, diarrhea, abdominal pain and headache. Arteritis, which
is characterized by blood vessel inflammation, hemorrhage, and
⇑
Corresponding author.
E-mail address: sandro.boland@amakem.com (S. Boland).
These authors contributed equally to the work.
Figure 1. Structure of the PDE4 inhibitors Rolipram and Roflumilast.
http://dx.doi.org/10.1016/j.bmcl.2014.07.016
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

S. Boland et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4594–4597
4595
selectivity has also been investigated, as PDE4B was proposed as
the main PDE4 isoform mediating TNF-
release,14 while knockout
experiments suggested that inhibition of PDE4D would be associ-
ated with emesis.15 However, the improved therapeutic window
observed with several PDE4 inhibitors during preclinical studies
was not always confirmed in clinical trials.8
oxidized to the corresponding benzoic acids 4a and 4b, using
sodium chlorite (Lindgren oxidation) in combination with sulfamic
acid as chlorine dioxide scavenger. Coupling with 3,5-dichloropyri-
din-4-amine, after in situ conversion of 4a and 4b into acid chlo-
rides, resulted in intermediates 5a and 5b. Further nucleophilic
substitution with either 3-amino- or 3-sulfanyldihydrofuranone
yielded final parent compounds 6–9 in a single step. The corre-
sponding metabolites 6M–9M could be easily obtained by hydroly-
sis of the lactone structure with LiOH. As N-oxide analogs of
Roflumilast and related PDE4 inhibitors were reported to possess
significant inhibitory activity against PDE4,11,25 compound 10, rep-
resenting the N-oxide analog of 6 was synthesized using the same
procedure as described above, by replacing 4-amino-3,5-dichloro-
pyridine with 4-amino-3,5-dichloropyridine-N-oxide in the corre-
sponding reaction step. Finally, compound 11, an acyclic analog
of 6 was synthesized by alkylating 5a with the methyl ester of gly-
cine. As with 6–9, the corresponding metabolite 11M was synthe-
sized via hydrolysis of the parent compound with LiOH.
Evaluation of the candidate PDE4 inhibitors revealed strong on-
target activity. Comparative data with Roflumilast and Rolipram
are given in Table 1. Parent compounds 6, 7, 8 and 9 displayed sim-
ilar IC50 values in the low nanomolar range. Replacement of the
pyridine moiety by a pyridine N-oxide resulted in a significant
drop in potency, as illustrated by 10. This finding was surprising,
as the N-oxide analogs of Roflumilast are known to possess signif-
icant inhibitory activity against PDE4.11,25 Compound 11, repre-
senting an acyclic analog of 6, displayed slightly reduced (2–4
fold) potency. Significant differences were found between the
expected metabolites 6M, 7M, 8M, 9M and 11M. Indeed, 6M and
8M, which resulted from hydrolysis of a homoserine lactone struc-
ture, were markedly less active than their parent compounds,
while their sulfanyl analogs 7M and 9M essentially retained their
on-target activity. This conserved activity was not the result of
re-lactonization of the metabolites under assay conditions or stor-
age in DMSO stock solutions, as both hypotheses were excluded by
HPLC–MS analysis. A major drop of potency was also found for
11M.
a
Another possibility is to optimize PDE4 inhibitors towards top-
ical delivery.8,16,17 In this context a soft drug approach represents
an attractive way towards novel PDE4 inhibitors with improved
therapeutic window. Soft drugs, sometimes known as antedrugs,
are biologically active compounds that are designed to undergo
metabolic inactivation by controlled conversion of the parent mol-
ecule into a predictable, nontoxic metabolite.18,19 This strategy has
recently been applied to a series of boron-containing PDE4 inhibi-
tors, in view of dermatologic applications.20 We have recently
reported the design and evaluation of soft inhibitors of ROCK, a
kinase target of relevance for the treatment of respiratory diseases,
including COPD.21 Herein, we discuss the design and initial evalu-
ation of soft PDE4 inhibitors which are analogs of Roflumilast.
Crystal structures of PDE4 in complex with several inhibitors,
including Roflumilast, have been reported and provide a structural
basis for their activity.22 Those structures revealed multiple hydro-
phobic interactions with highly conserved residues sandwiching
the inhibitor in the active site, as well as H-bonding interactions
between the (halo)alkoxy group found in several inhibitors and
an invariant glutamine. The lipophilic nature of the Roflumilast
binding site was further emphasized in a later publication wherein
a linker of sufficient length had to be included between the main
inhibitor scaffold and ionic solubilizing groups.5 Interestingly, such
findings suggest an inactivation mechanism for the design of soft
PDE4 inhibitors, since generation of a charged species by drug
metabolizing enzymes could result in a metabolite with decreased
potency. The associated change in physical–chemical properties
can also result in reduced membrane permeability, which repre-
sents another way of modulating the functional activity of com-
pounds inhibiting an intracellular target.
For the design of soft PDE4 inhibitors, we chose to favor a c-lac-
tone structure, which was previously used in soft corticosteroids23
and more recently in the design of soft ROCK inhibitors.24 Hydroly-
sis of this lactone will result in a negatively charged species, with
increased polar surface area and decreased hydrophobicity. Candi-
date soft PDE4 inhibitors were prepared in 5 or 6 steps, starting
from the commercially available 3,4-dihydroxybenzaldehyde 1
(Fig. 2). Sequential alkylation of 1 first involved selective difluorom-
ethylation on position 4, using sodium chlorodifluoroacetate as a
source of difluorocarbene. Alkylation of the remaining hydroxyl
function of 2 was performed with alkyl dibromides, yielding inter-
mediates 3a and 3b. In the next step, the aldehyde function was
An important part of the observed structure–activity relation-
ships can be explained by the charge state of the different species,
and by the distance between charged groups and the main inhibi-
tor scaffold. Indeed, it is known that highly polar moieties are
poorly tolerated in the essentially hydrophobic binding site of
PDE4, unless a spacer of sufficient length is placed between the
ionizable group and the main inhibitor scaffold.5 All parent com-
pounds remain neutral at physiological pH and under assay condi-
tions, and consequently display strong on-target activity. The fact
that metabolites from sulfanyl analogues 7M and 9M retain strong
on-target activity suggests that the carboxylic group resulting from
Figure 2. Overview of compound synthesis: (a) ClF2CCOONa, NaOH, DMF, H2O, 120 °C, 2 h; (b) Br(CH2)nBr, K2CO3, ACN, reflux, 4 h; (c) H2NSO3H, NaClO2, AcOH, H2O, 10 °C,
30 min; (d) (1) SOCl2, toluene, 90 °C, 1.5 h; (2) 3,5-dichloro-4-aminopyridine or 3,5-dichloro-4-aminopyridine N-oxide, 60% NaH, THF, 10 °C, 1 h; (e) 2-oxo-3-amino-oxolane,
2-oxo-3-mercapto-oxolane or glycine methyl ester; K2CO3, ACN, reflux, 1.5 h.

4596
S. Boland et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4594–4597
Table 1
would be found between the parent compounds and their
metabolites.
As homoserine lactone contains a chiral centre, the enantiomers
PDE4 inhibition data for synthesized compounds
of 6 were prepared, starting from commercially available, optically
pure building blocks. While stereochemistry did not influence
activity against PDE4B1, a slight difference of potency (7-fold)
between enantiomers was found when evaluating (R)-6 and (S)-6
against PDE4D2, in favor of the (R) enantiomer. Both of the corre-
sponding metabolites were markedly less active than their parent
compounds.
While 7M and 9M retain strong potency against PDE4, it should
be remembered that PDE4 is an intracellular target. Barring some
active transport into the cell, PDE4 inhibitors can therefore only
exert functional activity if they display some degree of membrane
permeability. Evaluation of the parent lactones with their corre-
sponding metabolites in a parallel artificial membrane permeabil-
ity assay (PAMPA) clearly indicates high apparent permeability
(Papp) for the four lactones 6–9 and the acyclic analog 11, with val-
ues comparable to propranolol (high Papp control). In contrast, their
respective metabolites all displayed a Papp value lower than raniti-
dine (low Papp control), thereby indicating a major (from 50 to 350-
fold) drop of Papp (Table 2). Those observations appear in line with
literature data, reporting on average lower permeability for acidic
and zwitterionic species compared to their neutral counterparts.26
This ‘permeability switch’ should in turn affect the functional
activity of the metabolites. For 6, 8 and 11, this means that the soft
drug approach is consolidated by a double inactivation mechanism,
Compds
X
Y
n
PDE4B1
IC50 (nM)
PDE4D2
IC50 (nM)
a
a
b
b
b
b
b
Rolipram
Roflumilast
6
—
—
—
N
N
N
N
N
N
N
N
N
N
N
N
N
—
—
3
3
3
3
3
3
3
3
4
4
4
4
3
3
3
52
130
0.7
18
7.3
48
2300
7200
>10,000
5.5
9.9
6.6
150
1.8
4.5
170
32
1500
0.4
5.4
12
NH
NH
NH
NH
NH
NH
S
(R)-6
(S)-6
6M
(R)-6M
(S)-6M
7
7M
8
8M
9
11
740
1700
2700
1.8
5.5
4.0
180
4.7
8.8
88
S
NH
NH
S
9M
10
11
S
NH
NH
NH
N+—Oꢀ
N
N
21
910
11M
namely
a combination of decreased on-target activity and
a
decreased permeability.
Values are means of two experiments.
Not applicable due to scaffold.
b
An important aspect of the soft drug approach resides in the
controlled conversion of the parent compound into its functionally
inactive metabolite. In this respect, the stability of the candidate
compounds in human plasma was assessed (Table 3). From these
data, it is evident that metabolism occurs readily, with half-life val-
ues (t1/2) below 20 min for parent compounds. Metabolites, on the
other hand, all appeared stable (t1/2 >60 min); suggesting no fur-
ther degradation in plasma. As potential applications of PDE4
inhibitors include respiratory diseases, selected compounds were
evaluated for stability in the presence of lung S9 fractions and dis-
played t1/2 >60 min. Finally, the stability of selected compounds in
the presence of homogenized human lung tissue (40 mg/ml) was
checked. Homoserine lactone derivatives 6 and 8 were found to
be more stable under tested conditions than 7. Those compounds
were equally stable in presence of mouse lung homogenate, and
can therefore be evaluated in rodent models of lung inflammation.
Further stability experiments were conducted in order to deter-
mine the enzyme(s) responsible for the rapid degradation in
plasma. Compound 6 and 7 was incubated in the presence of
various esterase inhibitors including phenylmethylsulfonyl
fluoride (PMSF, serine esterase and serine protease inhibitor),
hydrolysis of the c-lactone is too far from the main inhibitor scaf-
fold to influence on-target potency. However, pKa predictions
(Instant J Chem 6.1) suggest that while 6, 8 and 11 remain neutral
at physiological pH, their metabolites 6M, 8M and 11M should be
zwitterions (Fig. 3). The additional charged center that is present in
those metabolites is located closer to the main inhibitor scaffold
and is therefore able to affect on-target potency. The importance
of spacer length is further exemplified by the significantly higher
potency of metabolite 8M (vs. 6M), which displays a longer spacer.
At this stage, longer spacers were not investigated, since the gener-
ated data suggested that a lower difference of on-target activity
Table 2
PAMPA data for compounds 6–11 and their metabolites
Compds
Pampa Papp (10ꢀ6 cm/s)
Ranitidine
Propranolol
0.14
10
6
7.3
6M
7
0.02
6.6
7M
8
0.14
4.7
8M
9
0.03
2.5
9M
10
11
11M
0.05
ND
8.9
Figure 3. Hydrolysis of the lactone structure of 6 results in a zwitterionic species
6M at physiological pH.
<0.01

S. Boland et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4594–4597
4597
Table 3
this compound series resides in the characterization of their
in vivo efficacy in rodent models of lung inflammation. An assess-
ment of the emetic effects of 6 and 8 should also be made, in order
to fully demonstrate that soft PDE4 inhibitors can indeed provide
an increased therapeutic window.
Stability data for compounds 6–11
Compds
t1/2 plasmaa
(human)
t1/2 lungS9a
(human)
t1/2 lung hom.a
(human)
6
7
8
9
10
11
<5
<5
<5
<5
8
>60
>60
>60
>60
NT
>60
39
>60
NT
NT
NT
Acknowledgments
The authors wish to thank Dr. Jean Marie Stassen and Dr. Steve
Pakola for their contribution and advice during the review of this
manuscript.
19
NT
a
Half-life values in minutes.
Supplementary data
bis(4-nitrophenyl) phosphate (BNPP, carboxylesterase inhibitor),
tacrine (cholinesterase inhibitor) and EDTA (paraoxonase inhibi-
tor). Under most tested conditions (Fig. 4a), 6 consistently showed
very fast degradation (t1/2 around 2 min). However, addition of
EDTA was sufficient to completely prevent compound hydrolysis
over the course of the experiment (20 min). Such results were also
obtained with 7 and 8 and suggested that the tested compounds
are hydrolyzed in plasma exclusively through paraoxonase
(PON). This hypothesis was further tested by evaluating the stabil-
ity of 6 in the presence of recombinant, catalytically active PON1
enzymes representing the main PON1 variants in human (Q192
and R192). Degradation of 6 was clearly observed within minutes
in the presence of both PON1 variants (Fig. 4b).
In conclusion, a new series of locally active PDE4 inhibitors has
been prepared and their structure–activity relationship was inves-
tigated. Several compounds exhibited low nanomolar on-target
activity against PDE4B1 and PDE4D2. From the results above, 6
and 8 emerged as the most promising candidates, combining low
nanomolar activity with rapid degradation (t1/2 <5 min) into a
functionally inactive metabolite in plasma and good stability in
lung tissues. The next step in the evaluation and development of
Supplementary data (experimental protocols regarding synthe-
sis, characterization and evaluation of compounds) associated with
this article can be found, in the online version, at http://dx.doi.org/
10.1016/j.bmcl.2014.07.016.
References and notes
1. Souness, J. E.; Aldous, D.; Sargent, C. Immunopharmacology 2000, 47, 127.
2. Conti, M.; Jin, S. L. Prog. Nucleic Acid Res. Mol. Biol. 1999, 63, 1.
3. Soto, F. J.; Hanania, N. A. Curr. Opin. Pulm. Med. 2005, 11, 129.
4. Lipworth, B. J. Lancet 2005, 365, 167.
5. Govek, S. P.; Oshiro, G.; Anzola, J. V.; Beauregard, C.; Chen, J.; Coyle, A. R.;
Gamache, D. A.; Hellberg, M. R.; Hsien, J. N.; Lerch, J. M.; Liao, J. C.; Malecha, J.
W.; Staszewski, L. M.; Thomas, D. J.; Yanni, J. M.; Noble, S. A.; Shiau, A. K. Bioorg.
Med. Chem. Lett. 2010, 20, 2928.
6. Bäumer, W.; Hoppmann, J.; Rundfeldt, C.; Kietzmann, M. Inflamm. Allergy Drug
Targets 2006, 6, 17.
7. Salari, P.; Abdollahi, M. Expert Opin. Invest. Drug 2012, 21, 261.
8. Pagès, L.; Gavaldà, A.; Lehner, M. D. Expert Opin. Ther. Pat. 2009, 19, 1501.
9. Gavaldà, A.; Roberts, R. S. Expert Opin. Ther. Pat. 2013, 23, 997.
10. Hirose, R.; Manabe, H.; Nonaka, H.; Yanagawa, K.; Akuta, K.; Sato, S.; Ohshima,
E.; Ichimura, M. Eur. J. Pharmacol. 2007, 573, 93.
11. Tashkin, D. P. Expert Opin. Pharmacol. 2014, 15, 85–96.
12. Poole, R. M.; Ballantyne, A. D. Drugs 2014, 74, 825.
13. Torphy, T. J.; Barnette, M. S.; Underwood, D. C.; Griswold, D. E.; Christensen, S.
B.; Murdoch, R. D.; Nieman, R. B.; Compton, C. H. Pulm. Pharmacol. Ther. 1999,
12, 131.
14. Jin, S. L.; Conti, M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 7628.
15. Robichaud, A.; Stamatiou, P. B.; Jin, S. L.; Lachance, N.; MacDonald, D.; Laliberte,
F.; Liu, S.; Huang, Z.; Conti, M.; Chan, C. C. J. Clin. Invest. 2002, 110, 1045.
16. Woodrow, M. D.; Ballantine, S. P.; Barker, M. D.; Clarke, B. J.; Dawson, J.; Dean,
T. W.; Delves, C. J.; Evans, B.; Gough, S. L.; Guntrip, S. B.; Holman, S.; Holmes, D.
S.; Kranz, M.; Lindvaal, M. K.; Lucas, F. S.; Neu, M.; Ranshaw, L. E.; Solanke, Y. E.;
Somers, D. O.; Ward, P.; Wiseman, J. O. Bioorg. Med. Chem. Lett. 2009, 19, 5261.
17. Armani, E.; Amari, G.; Rizzi, A.; De Fanti, R.; Ghidini, E.; Capaldi, C.; Carzaniga,
L.; Caruso, P.; Guala, M.; Peretto, I.; La Porta, E.; Bolzoni, P. T.; Facchinetti, F.;
Carnini, C.; Moretto, N.; Patacchini, R.; Bassani, F.; Cenacchi, V.; Volta, R.;
Amadei, F.; Capacchi, S.; Delcanale, M.; Puccini, P.; Catinella, S.; Civelli, M.;
Villetti, G. J. Med. Chem. 2014, 57, 793.
18. Bodor, N.; Buchwald, P. In Burger’s Medicinal Chemistry, Drug Discovery and
Development; John Wiley & Sons: New York, 2003; pp 533–608. 6th ed.
19. Erhardt, P. W.; Reese, M. D. In Prodrugs and Targeted Delivery: Towards Better
ADME Properties; Rautio, J., Ed.; Wiley-VCH: Weinheim, 2011; pp 385–413.
20. Zhang, Y. K.; Plattner, J. J.; Akama, T.; Baker, S. J.; Hernandez, V. S.; Sanders, V.;
Freund, Y.; Kimura, R.; Bu, W.; Hold, K. M.; Lu, X. S. Bioorg. Med. Chem. Lett.
2010, 20, 2270.
21. Boland, S.; Defert, O.; Alen, J.; Bourin, A.; Castermans, K.; Kindt, N.; Boumans,
N.; Panitti, L.; Van de Velde, S.; Stalmans, I.; Leysen, D. Bioorg. Med. Chem. Lett.
2013, 23, 6442.
22. Card, G. L.; England, B. P.; Suzuki, Y.; Fong, D.; Powell, B.; Lee, B.; Luu, C.;
Tabrizizad, M.; Gillette, S.; Ibrahim, P. N.; Artis, D. R.; Bollag, G.; Milburn, M. V.;
Kim, S.-H.; Schlessinger, J.; Zhang, K. Y. J. Structure 2004, 12, 2233.
23. Biggadike, K.; Angell, R. M.; Burgess, C. M.; Farrell, R. M.; Hancock, A. P.; Harker, A.
J.; Irving, W. R.; Ioannou, C.; Procopiou, P. A.; Shaw, R. E. J. Med. Chem. 2000, 43, 19.
24. Leysen, D.; Defert, O.; Boland, S. WO/2013/030216, 2013.
25. Bethke, T. D.; Böhmer, G. M.; Hermann, R.; Hauns, B.; Fux, R.; Mörike, K.; David,
M.; Knoerzer, D.; Wurst, W.; Gleiter, C. H. J. Clin. Pharmacol. 2007, 47, 26.
26. Gleeson, M. P. J. Med. Chem. 2008, 51, 817.
Figure 4. (a) Effect of esterase inhibitors on the hydrolysis of 6 in human plasma.
(b) Hydrolysis of 6 by recombinant human PON1 variants.