Late-Stage Lead Diversification Coupled with Quantitative Nuclear Magnetic Resonance Spectroscopy to Identify New Structure-Activity Relationship Vectors at Nanomole-Scale Synthesis: Application to Loratadine, a Human Histamine H1Receptor Inverse Agonist

Article abstract of DOI:10.1021/acs.jmedchem.0c00483

An experimental approach is described for late-stage lead diversification of frontrunner drug candidates using nanomole-scale amounts of lead compounds for structure-activity relationship development. The process utilizes C-H bond activation methods to explore chemical space by transforming candidates into newly functionalized leads. A key to success is the utilization of microcryoprobe nuclear magnetic resonance (NMR) spectroscopy, which permits the use of low amounts of lead compounds (1-5 μmol). The approach delivers multiple analogues from a single lead at nanomole-scale amounts as DMSO-d6 stock solutions with a known structure and concentration for in vitro pharmacology and absorption, distribution, metabolism, and excretion testing. To demonstrate the feasibility of this approach, we have used the antihistamine agent loratadine (1). Twenty-six analogues of loratadine were isolated and fully characterized by NMR. Informative SAR analogues were identified, which display potent affinity for the human histamine H1 receptor and improved metabolic stability.

Full text of DOI:10.1021/acs.jmedchem.0c00483

Subscriber access provided by ECU Libraries
Article
Late Stage Lead Diversification (LSLD) Coupled with Quantitative NMR
Spectroscopy to Identify New SAR Vectors at Nanomole Scale Synthesis:
1
Application to Loratadine a Human Histamine H Receptor Inverse Agonist
Manjinder S. Lall, Asser Bassyouni, James Bradow, Maria Brown, Mark W Bundesmann,
Jinshan Chen, Gregory Ciszewski, Anne Hagen, Dennis Hyek, Stephen Jenkinson,
Bo Liu, R. Scott Obach, Senliang Pan, Usa Reilly, Neal W. Sach, Daniel Smaltz,
Douglas K. Spracklin, Jeremy Starr, Melissa Wagenaar, and Gregory S Walker
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00483 • Publication Date (Web): 28 May 2020
Downloaded from pubs.acs.org on May 28, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Late Stage Lead Diversification (LSLD) Coupled with Quantitative NMR Spectroscopy to
Identify New SAR Vectors at Nanomole Scale Synthesis: Application to Loratadine a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Human Histamine H Receptor Inverse Agonist
1
1
2
1
1
1
Manjinder S. Lall,* Asser Bassyouni, James Bradow, Maria Brown, Mark Bundesmann,
1
1
1
3
2
3
Jinshan Chen, Greg Ciszewski, Anne E. Hagen, Dennis Hyek, Stephen Jenkinson, Bo Liu, R.
1
1
1
2
4
1
Scott Obach, Senliang Pan, Usa Reilly, Neal Sach, Daniel J. Smaltz, Douglas K. Spracklin,
1
1
1
Jeremy Starr, Melissa Wagenaar, Gregory S. Walker
Corresponding Author:
E-Mail: manjinder.lall@pfizer.com
*
1Pfizer Worldwide Research and Development, Eastern Point Road, Groton, Connecticut 06340,
United States
2Pfizer Worldwide Research and Development, Science Center Drive, San Diego, California
9
2121, United States
3Spectrix Analytical Services, LLC, 410 Sackett Point Road, Bldg 20, North Haven, Connecticut
0
6473, United States
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4Pfizer Worldwide Research and Development, 1 Portland Street, Cambridge, Massachusetts
0
2139, United States
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
2

Page 3 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ABSTRACT: An experimental approach is described for late stage lead diversification (LSLD)
of frontrunner drug candidates using nanomole scale amounts of lead compounds for SAR
development. The process utilizes C–H bond activation methods to explore chemical space by
transforming candidates into newly functionalized leads. A key to success is the utilization of
micro-cryoprobe NMR spectroscopy which permits the use of low amounts of lead compounds (1-
5 moles). The approach delivers multiple analogs from a single lead at nanomole scale amounts
as DMSO-d6 stock solutions with known structure and concentration for in vitro pharmacology
and ADME testing. To demonstrate the feasibility of this approach, we have used the
antihistamine agent loratadine (1). Twenty-six analogs of loratadine were isolated and fully
characterized by NMR. Informative SAR analogs were identified which display potent affinity
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
for the human histamine H receptor and improved metabolic stability.
1
KEYWORDS: late stage lead diversification (LSLD)
diversity focused synthesis (DFS) positive/negative structure-activity relationship (+/-SAR)
miniaturization, nanoscale drug discovery C–H bond activation liver microsomes
cytochrome P450 microorganisms metalloporphyrins Minisci reaction
Baran Diversinates photoredox catalysis
MS/MS) pharmacology testing

late stage functionalization (LSF)











electrochemistry



micro-cryoprobe NMR tandem mass spectrometry

(


loratadine
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Late stage lead diversification (LSLD) is emerging as a powerful synthetic strategy for drug
1
discovery. This approach leverages unactivated C–H bonds as sites to incorporate functional
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
groups or small substituents late in the synthetic route of a lead molecule by applying biochemical
2
and chemical functionalization methods (Figure 1A). From a medicinal chemist’s perspective,
lead diversification is a highly attractive approach because it does not require chemical pre-
functionalization of starting materials or de novo synthesis for each new analog targeted.
Furthermore, because pharmacological activity is already present in the lead molecule LSLD has
the ability to rapidly and cost-effectively incorporate drug-like properties and attributes into
promising candidates. Allowing the medicinal chemist to expand structure-activity relationship
(SAR) knowledge, optimize ligand-target potency and selectivity, alter absorption-distribution-
metabolism-excretion (ADME) properties, improve physicochemical properties like solubility,
identify metabolically labile sites to aid drug design, access new and synthetically challenging
unexplored chemical space and uncover novel intellectual property (IP) space. Therefore, this
approach can significantly accelerate the drug discovery process with minimal synthetic
investment, time and cost.
Drug discovery and development is challenging because most drugs are designed through
iterative rounds of chemical synthesis and pharmacological testing to incorporate drug-like
3
attributes into lead molecules. The LSLD approach is appealing given that drug-like chemical
space has been estimated at 1060 for molecules obeying Lipinski’s rule-of-five for oral
2
0
24
4
bioavailability and at 10 -10 for molecules up to 30 atoms. To manage these formidable
numbers, LSLD lends itself to methodology that favors diversity focused synthesis (DFS) where
the aim is to explore a dense area of chemical space adjacent to regions known to have useful
ACS Paragon Plus Environment
4

Page 5 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
properties, such as biological activity (Figure 1B). Hence, library size is not as important as
6
libraries with focused diversity with favorable physicochemical properties (e.g. LipE) for a
particular biological target.7
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. A) Late stage lead diversification (LSLD) to generate lead diversity in an efficient
manner for a specific biological target. B) Diversity focused synthesis (DFS) explores related areas
of chemical space in proximity to regions known to have useful properties.
Much attention has recently been focused on late stage C–H bond activation and
8
functionalization methods to diversify lead molecules in drug discovery. Incorporating functional
groups or small substituents efficiently with judicious range in size and polarity into drug-like
molecules is highly relevant for SAR development, can accelerate analog preparation and could
potentially reduce high attrition associated with pharmaceutical research.9a For example, high
lipophilicity of drug candidates can lead to poor oral absorption, low compound solubility and high
metabolic turnover.9b Late stage C–H bond modification of biologically active compounds is an
effective approach to address these issues because direct replacement of C–H bonds with
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
functional groups or substituents can significantly improve target binding or favorably influence
receptor selectivity. Previously, we have described biosynthetic methods to carry out lead
diversification and drug metabolite synthesis coupled with quantitative micro-cryoprobe NMR
spectroscopy1b-d,10 which laid the foundation for the present work. Herein, we report the
operationalization and expansion of an approach to facilitate late stage lead derivatization and
metabolite synthesis for advanced lead molecules using small amounts (i.e. 1-2 mg) of compound
to support drug discovery. The approach generates chemical diversity of lead molecules through
an alternative approach to classical medicinal chemistry practices, for rapid SAR knowledge and
chemical space exploration. In particular, the approach utilizes direct C–H bond functionalization
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
methods in support of diversity focused synthesis to transform candidates into newly substituted
leads or noteworthy metabolites, such as: mammalian liver microsomes and cytosols;1b-d
microorganisms;11 engineered and recombinant P450 enzymes;10,12 biomimetic
metalloporphyrins;13 electrochemistry;14 Minisci chemistry;15 alkyl sulfinate radical precursor
16
17
chemistry (Baran Diversinates); and photoredox catalysis. Therefore, the approach enables 1)
synthetic modification of lead compounds to improve biological properties; and 2) synthesis of
metabolites for in vitro profiling to inform intrinsic clearance (CL ) parameters. Key to our
int
method is the underpinning analytical infrastructure which uses low amounts of lead compounds
to carry out screens for hit identification, scale-up of interesting products, purification using liquid
chromatography with automated fraction collection for isolation, full structure characterization
based on high resolution mass spectrometry plus tandem mass spectrometry (MS/MS) and NMR
spectroscopy utilized for both structure elucidation and quantitation (Figure 2). With little
investment up-front (1-2 mg of lead), the approach delivers multiple analogs from a single lead at
nanomole scale amounts (~5-10 g per product) as DMSO-d6 stock solutions with known structure
ACS Paragon Plus Environment
6

Page 7 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and concentration for immediate pharmacology and ADME testing. The approach can rapidly and
cost-effectively address multiple medicinal chemistry problems on miniaturized scale, such as:
provide valuable +/-SAR information, improve potency and selectivity, identify metabolically
labile sites to aid drug design, alter physicochemical properties and access novel synthetically
challenging chemical space. To demonstrate our approach, we have generated compound diversity
and drug metabolites using the antihistamine agent loratadine (1) as a prototypical compound to
traverse through our workflow with the aim to explore chemical space about the A-D ring systems
and identify new SAR vectors.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Workflow for late stage lead diversification.
RESULTS AND DISCUSSION
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The clinical agent loratadine is a second generation antihistamine drug used to treat allergic
rhinitis, also known as hay fever.18 Well-known for more than 20 years, loratadine has been
subject to extensive SAR.19a-d Loratadine is a tricyclic non-sedating inverse agonist of the human
histamine H receptor (dissociation constant, K = 0.19 M) and is preferred over first generation
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
d
antihistamines because of lack of sedation and CNS adverse effects.19e-g Loratadine was used as
an example compound to test our approach and highlight the ability to generate diverse compounds
because the molecule contains drug-like structural features common to typical lead molecules
explored during drug discovery, such as sites containing: alkyl, O/N-alkyl, aryl, heteroaryl,
carbonyl (carbamate), haloarene and unsaturation functionality (Figure 3). To explore loratadine
SAR, we envisioned using various C–H bond activation methods to synthetically access and
functionalize the A-D ring systems in 1. The total synthesis of loratadine is complex and involves
7
steps with an overall yield of 37%, after considerable process development.19a,b Preparing SAR
analogs of 1 with systematic and probing modification at all ring systems (A-D) in single
transformative steps is challenging but conceivable using our approach.
Figure 3. Functional group distribution of antihistamine drug loratadine (1) and lead
diversification technologies employed for LSLD to access A-D ring system SAR.
ACS Paragon Plus Environment
8

Page 9 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Liver Microsomes and Recombinant CYP Enzymes: To diversify the alkyl sites in loratadine
(A- and C-rings), compound 1 was screened against a panel of mammalian liver microsomes (9
systems) and human recombinant cytochrome P450 enzymes (9 CYP isoforms), configured in 96-
well microtiter plate format for automated screening convenience as described in Table S1 in the
Supporting Information section. Despite the large number of C–H bonds in loratadine, CYPs
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
possess the exceptional capability to react at inert C(sp )–H bonds within carbon backbones
preferentially over other more chemically reactive sites. Hence, cytochrome P450 enzymes are
versatile chemo- and regioselective biocatalysts and are known to catalyze a wide range of
reactions including but not exclusive to hydroxylation, O- and N-dealkylation, heteroatom
oxidation, dehydrogenation and epoxidation.20 Some of these transformations have the unique
ability to introduce polarity (e.g. hydroxylation) at synthetically challenging sites and significantly
2
lower lipophilicity; a single hydroxyl group adds 20 Å of polar surface area (PSA) which can
often lead to increased metabolic stability.21 Furthermore, the hydroxyl group can behave as a
hydrogen bond acceptor (HBA) or hydrogen bond donor (HBD) providing binding energy to
biological targets resulting in increased ligand affinity. With these multifaceted capabilities, this
functional group offers medicinal chemists a structural change that can modify a lead molecules
behavior at the target site and in the body.21b As a result, the hydroxyl group is present in many
biologically important molecules including ~20% of drugs22a,b and ~65% of natural products.22c
Accordingly, chromatographic profiles from liver microsomes and recombinant CYP
screens (Table S2 in the Supporting Information) were examined by UV and tandem mass
spectrometry (MS/MS) and the results evaluated for acceptable product yield (>10% conversion)
and product diversity. It is important to mention that percent conversion to end-products as well
as diversity of products are critical factors when considering which system should be selected for
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
scale-up. Also, reaction profiles which provide products in the molecular weight range of the
starting material are favored over those which generate lower molecular weight product fragments
via heteroatom dealkylation reactions. For loratadine, thirteen products were detected by HPLC-
MS (Figure 4) where M+/-number refers to the average mass change difference between parent
and analog (i.e. delta mass). The HPLC-MS results from the liver microsome and recombinant
CYP screen show wells 9 and 18 generating good diversity, affording up to ten unique products
with reasonable conversion. Products generated from wells 9 and 18 correspond to incubations
with human liver microsomes and recombinant CYP3A4 (respectively) which is in good
agreement with previous reports that show CYP3A4 and CYP2D6 as the predominant enzymes
involved in loratadine in vitro metabolism.23
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
0

Page 11 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Liver microsomes and recombinant CYP HPLC-MS chromatograms for human liver
microsomes (well 9), rhCYP3A4 (well 18) and rabbit liver microsomes (well 6). M+/-number
refers to the average mass change difference between parent and analog (i.e. delta mass).
Screening lead molecules using liver microsomes in concert with recombinant CYPs, as in
Table S1, also provides a prospective opportunity to forecast CYP liability and determine which
isoforms are most active in vitro; offering an early glimpse into metabolite profiling and
identification (Met ID).24 Furthermore, we have established protein-ligand docking protocols
based on known CYP isoform X-ray structures to guide potential binding modes of leads with
CYP active sites using binding energy and shape complementarity scores.25 Caution should be
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
used when utilizing protein-ligand docking to explain metabolism because of the highly dynamic
nature of the P450 enzyme binding pocket. Thus, these initiatives are predictive but can potentially
address drug design approaches to enhance metabolic stability given that a high rate of metabolism
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
i.e. high intrinsic clearance, CL ) is a function of enzyme maximum velocity (Vmax) and the
int
Michaelis constant (Km) (Figure 5).26a In view of this, metabolic clearance by P450s can be
impacted through lead diversification strategies that block sites of metabolism, alter ligand
electronics or introduce polarity into lead molecules.26b
A) CYP2D6 Model with Loratadine
Loratadine
°
4
A
Iron(IV)-oxo
porphyrin
B)
Potential strategies to impact CLint
:
1
) Decrease Vmax
Vmax
Km

Block site of metabolism on lead
Alter lead electronics
CLint
=

2) Increase Km

Decrease enzymes affinity for lead
Figure 5. A) Modeled pose of loratadine with human CYP2D6. The oxygen atom of the CYP2D6
iron(IV)-oxo porphyrin radical cation is 4 Å from one of the methylene hydrogen atoms of the
ACS Paragon Plus Environment
1
2

Page 13 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ethyl carbamate moiety in loratadine. B) Potential strategies to alter a lead molecule’s intrinsic
clearance (CL ) value.
int
Microorganisms: Microbes have long been an important source of bioactive natural products for
drug discovery, providing a wealth of novel scaffolds as well as marketed drugs.11,27 Most
recently, microbial incubations with lead molecules have been investigated to chemically diversify
drug candidates to expand SAR knowledge or incorporate chemical handles for further structural
modification.28 Other virtues of microbial biotransformations include their ability to mimic
mammalian metabolism or produce microbe specific metabolites. They often produce multiple
products, can generate compounds difficult to produce through synthetic chemistry, are scalable
processes (mg to kg quantities), produce oxidative (phase I) and conjugative (phase II) metabolites
and are cost-effective.29 Thus, loratadine was screened for modification on ring systems A and C
using a panel of 79 microorganisms which included organisms from bacteria, fungi and
actinomycetes (Table S3 in the Supporting Information).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Loratadine was incubated with microbial panel (Table S3) to afford twenty-three products
observed by HPLC-MS (Table S4 in the Supporting Information). As expected when using live
whole cells, the microbial screen gave many products from phase I and II metabolism of loratadine.
In addition, we observed non-drug related secondary metabolite products (e.g. M+63, M+65,
M+84 and 2M+96) which can be differentiated from 1 by MS/MS fragmentation patterns. Of the
twenty-three compounds detected, eighteen products appeared to be drug related by HPLC-MS
(
Figure 6).30a In considering the chromatographic profiles generated for loratadine using microbial
whole cells, it was observed that bacterial organisms (e.g. well 19, Nocardia ATCC 21271a: 10
products detected) were more efficient at generating diversity as compared to fungal strains (e.g.
well 65, Mucor rouxii UC7470: 6 products detected).30b
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Microorganisms HPLC-MS chromatograms for Nocardia ATCC 21271a (well 19),
Mucor rouxii UC7470 (well 65) and Streptomyces fradiae (well 15).30a M+/-number refers to the
average mass change difference between parent and analog (i.e. delta mass).
Engineered Cytochrome P450 Enzymes: Mammalian cytochrome P450 (CYP) enzymes are a
superfamily of membrane-bound heme containing monooxygenases that can insert atmospheric
3
oxygen into unactivated C(sp )–H bonds across an enormous variety of substrates under mild
3
1
conditions with a high degree of chemo- and regioselectivity. Despite these noteworthy qualities,
mammalian CYP applications are limited because of low turnover rate, enzyme instability and
they acquire electrons through a complex process involving cytochrome P450 reductase and
3
2
NADPH. For these reasons, there is considerable interest in utilizing directed evolution to
ACS Paragon Plus Environment
1
4

Page 15 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
engineer bacterial cytochrome P450 enzymes for the synthesis of novel drugs and drug
metabolites.33
Bacterial cytochrome P450 enzymes can be evolved to afford new biocatalysts with
enhanced synthetic applications to generate unique products not observed by mammalian CYPs
and provide valuable +/-SAR for drug discovery. Cytochrome P450 from Bacillus megaterium
CYP102A1 (P450 BM3) has characteristics that enable protein engineering for use in synthesis
because it is expressed in E. coli at high levels, it is a soluble enzyme and its electron transfer and
reductase domains are fused in one continuous polypeptide chain to efficiently facilitate oxidative
transformations.12,34 In this respect, tailoring the functionality of P450 BM3 has enabled its utility
as an effective biocatalytic tool in pharmaceutical research with wide-ranging synthetic
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
applicability from early drug discovery on small scale35 to late development on kilogram scale.36
With these salient features in mind, we utilized a cytochrome P450 BM3 panel comprised of 45
3
7
Codexis variants selected for their ability to cover the reactivity scope of human CYPs by
producing mammalian drug metabolites and also to show broad specificity on drug-like substrates
to provide new diverse chemical leads (Table S5 in the Supporting Information).
To this end, loratadine was screened against our P450 BM3 panel and evaluated for enzyme
functionalization activity and variety in product distribution about the A-, C- and D-ring systems.
As expected, the reactivity scope of the 45-membered P450 BM3 panel was broad and produced
a variety of products when applied against loratadine (Table S6 in Supporting Information),
providing a product profile that complemented the mammalian liver microsome and recombinant
CYP screen by generating different products. Seventeen different products were identified in the
HPLC-MS chromatograms of select P450 BM3 reactions with 1 as substrate (Figure 7). The
chromatographic traces from wells 40 (MCYP0012) and 47 (MCYP0160) show good diversity
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
profiles, generating up to sixteen products. A few of the P450 BM3 derived products matched
retention times and MS/MS fragmentation patterns to those which arose from mammalian CYPs,
which speaks to the panel’s capability to prepare authentic drug metabolites. In addition, several
BM3 variants gave products not observed when screening liver microsomes and recombinant
CYPs underscoring their utility for lead diversification. Several BM3 enzymes displayed good
degrees of regioselective-hydroxylation as exhibited by yielding five chromatographically
different M+16 products (vide infra). These enzymes which give rise to single hydroxylation
products are vital to SAR development because they effectively probe the lead molecule’s ability
to tolerate polar functionality at various sites, aiding drug design strategies for improved properties
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
8
(e.g. target potency, solubility, oral bioavailability, etc.). Furthermore, there was low occurrence
of bis-hydroxylation which reflects the enzyme’s exquisite selectivity for 1 over its more polar
hydroxylated products limiting the synthesis of unwanted poly-hydroxylated compounds.
ACS Paragon Plus Environment
1
6

Page 17 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. BM3 variant HPLC-MS chromatograms for MCYP0160 (well 47), MCYP0012 (well
4
0) and MCYP-P1.2-C11 (well 29). M+/-number refers to the average mass change difference
between parent and analog (i.e. delta mass).
Electrochemistry: Electrochemical transformations are a simple and green method for enabling
C–H bond functionalization of lead molecules, where addition or removal of electrons across an
electrical potential can bring about lead modification or metabolite synthesis.14 The advantage of
electrochemistry is the ability to precisely control both the potential and rate of the redox process
in a manner not possible with conventional chemical oxidants or reductants. For example,
electrochemical functionalization has been used to prepare derivatives of the antibiotic tylosin
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
where modification of the parent compound gives rise to novel desmycosin analogs.39
Electrochemical oxidation is also a useful tool to understand and study metabolic pathways, as
demonstrated for the antimalarial drug amodiaquine.40 To accommodate our needs we used a
NuVant EZStat Pro potentiostat to acquire cyclic voltammetry (CV) readings to determine redox
properties. Of particular interest is the initial anode oxidation potential which provides voltage
parameters for the subsequent electro-oxidation reaction. As a result, a preparative undivided
electrochemical cell was used to screen 1 (Figure S1A in the Supporting Information). Loratadine
was subjected to electrolysis with constant potential (5 V) across the two electrodes to modify the
olefinic region which flanks rings A and C in the molecule.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
After electrochemical initiation, the reaction progress was monitored by HPLC-MS to
follow formation of products. The reaction profile showed the formation of ten distinct products
(Table S7 in the Supporting Information). The observed electrochemical products appeared unique
by HPLC-MS (Figure 8) compared to products derived from biological oxidation (i.e. liver
microsomes, recombinant CYP, microorganisms and BM3). This illustrates differences between
the transformational methods and shows good complementarity between chemical and biological
oxidative technologies to deliver a diverse set of products.
ACS Paragon Plus Environment
1
8

Page 19 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. Electrochemistry HPLC-MS chromatogram. Electrolysis reaction conditions: graphite
anode, stainless steel cathode, Et N-BF , 5 V, MeCN / H O, RT, 1 hr. M+/-number refers to the
4
4
2
average mass change difference between parent and analog (i.e. delta mass).
Biomimetic Metalloporphyrin Oxidation: Synthetic metalloporphyrins capable of mimicking
the selective C–H bond oxidation of heme containing monooxygenase cytochrome P450s have
become of increasing significance to drug discovery in recent years.41 These biomimetic
metalloporphyrin oxidation (BMO) catalysts have been extensively investigated for lead
diversification and metabolite synthesis.42 Consequently, this method has been utilized for
4
3
aliphatic hydroxylation, O- and N-dealkylation, S- and N-oxidation and epoxidation, in addition
44
to other transformations. In view of this, loratadine was screened against a 120-membered BMO
panel consisting of two solvent systems (MeOH / DCE or MeCN / DCE), six oxidants (PhIO,
H O , Oxone, CHP / imidazole, mCPBA, or CHP / imidazole / HCO H) and nine commercially
2
2
2
available metalloporphyrins45 ((Fe(TNO PP), Mn(TFPP), Fe(TFPP), Cu(TFPP), Co(TFPP),
2
Pd(TDCIPP), Mn(TDCIPP), Cu(TDCIPP) or Co(TDCIPP)) (Table S8 in the Supporting
Information). The requisite screen was run under protic (MeOH / DCE, 1:1) and aprotic (MeCN
/ DCE, 1:1) solvent conditions, in an attempt to diversify the olefinic portion of loratadine and the
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
D-ring system. To ensure metalloporphyrin was actively participating in product formation a
control screen was run with solvent and oxidant, but no metalloporphyrin.
Chromatographic profiles from the BMO screen were evaluated by HPLC-MS and showed
the formation of six products (Table S9 in the Supporting Information). Once again the products
derived from the BMO screen appeared distinct by retention time and MS/MS fragmentation
pattern (Figure 9) from products derived from biological oxidation, indicating good
complementarity between chemical and biological oxidation techniques to enable lead
diversification.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 9. Biomimetic metalloporphyrin oxidation HPLC-MS chromatograms. Top
chromatogram: reaction conditions with metalloporphyrin Cu(TDCIPP), cumene hydroperoxide,
ACS Paragon Plus Environment
2
0

Page 21 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MeOH, DCM. Bottom chromatogram: reaction conditions without metalloporphyrin, mCPBA,
MeOH, DCM. M+/-number refers to the average mass change difference between parent and
analog (i.e. delta mass).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Comparison of Lead Diversification Screens. With the HPLC-MS results in hand from all
screens, we were able to compare and contrast chromatographic profiles from the different lead
diversification and metabolite synthesis technology screens. Early on, it became apparent that
chromatographic profiles derived from biological oxidative methods (i.e. liver microsomes,
recombinant CYP, microorganisms and BM3) gave similar product profiles by LC retention time
and MS/MS fragmentation, albeit with different product conversions (Figure 4, 6 and 7) which is
important for scale-up considerations. These results were in stark contrast to chromatographic
profiles obtained from EC and BMO screens (Figure 8 and 9), which gave relatively distinct
products. Only one product, as determined by LCMS analysis, was generated in both the EC and
BMO screens (e.g. m/z 399j, M+16) (Figure 9). Furthermore, the comparable m/z 399j product
was generated in the control BMO well without metalloporphyrin suggesting loratadine is
modified by oxidant mCPBA and not metalloporphyrin.
We also compared product similarity and difference relative to each biological technology
by using a diversity map where screening panels from all biological systems (e.g. liver
microsomes, recombinant CYP, microorganisms and BM3) are plotted as a function of product
detected by HPLC-MS (Figure 10, see Excel S1_Diversity Map in the Supporting Information
section to view entire plot). Viewing the data in this manner allowed for systematic identification
of transformational systems which gave comparable products and, more importantly, to pinpoint
systems which generate good product diversity.
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 10. Diversity map of product similarity and difference relative to biological technology
screened by HPLC-MS, refer to (Excel S1_Diversity Map) in the Supporting Information to view
entire map. X-axis displays product distribution determined by HPLC-MS, shown in yellow (top);
Y-axis displays well distribution from each biological screening panel: liver microsomes and
recombinant CYP (shown in brown), microorganisms (shown in green) and BM3 variants (shown
in red).
For example, extensive biotransformation of loratadine is observed from the diversity map using
human liver microsomes (well 9) which generates nine products by HPLC-MS (m/z 311 (M-72),
3
81c (M-2), 381d (M-2), 397a (M+14), 397b (M+14), 399a (M+16), 399b (M+16), 415a (M+32)
and 415b (M+32)). In a complimentary manner, the map also reveals four of these products (m/z
81c (M-2), 381d (M-2), 399a (M+16) and 399b (M+16)) are also made by microorganism
3
Streptomyces violascens ATCC 31560 (well 27) and BM3 variant MCYP0012 (well 40). Figure
10 also uncovers that products (m/z 327a (M-56), 327b (M-56) and 327c (M-56)) are generated by
rabbit liver microsomes (well 6), product (m/z 413c (M+32)) is made by microbe Streptomyces
4
6
setae ATCC 33774 (well 28), and products (m/z 397f (M+14), 397g (M+14), 413b (M+30) and
4
15e (M+32)) are generated by BM3 variant MCYP0012 (well 40).
ACS Paragon Plus Environment
2
2

Page 23 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We further summarized the results from the diversity map to illustrate the distribution of
similar products generated from the biological screens. For instance, liver microsome and
recombinant CYP products observed by HPLC-MS were compared to the those generated by other
biological technology screens e.g. microorganisms and BM3 (Figure 11). The results illustrated
in Figure 11 show loratadine products generated from liver microsomes and recombinant CYPs
have 69% product similarity to products generated from the microbial panel, 46% product
similarity to products generated from the BM3 panel and 46% similarity to all panels. For a more
comprehensive breakdown of product similarity relative to the other technologies screened by
HPLC-MS refer to Table S10 in the Supporting Information.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 11. Ratio of liver microsome and recombinant CYP products observed by HPLC-MS
relative to similar products detected from other technology screens (e.g. microbes and BM3). Venn
diagram representation (bottom).
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Lead Diversification Product Scale-Ups. With the diversity map in place, we shifted our
attention to using the plot to select wells for product scale-up with a focus on lead diversification
of loratadine with the specific aim to explore SAR chemical space by isolating products with
modifications on rings A-D. For instance, Figure 10 shows rabbit liver microsomes (well 6)
capable of converting loratadine to three unique products (m/z 327a-c, M-56) which are not
observed in any of the other technology screens. Therefore, male rabbit liver microsomes (New
Zealand, Xenotech) were selected for product scale-up which involved scaling the original liver
microsomes screen conditions by a factor of 80-fold at a protein concentration of 2 mg/mL and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
o
substrate (1) concentration of 20.0 M. Consequently, loratadine was incubated at 37 C utilizing
a reactor open to the atmosphere. After 1 hour, the reaction mixture was quenched with
acetonitrile, centrifuged and adsorbed onto a C18 HPLC column. The column was transferred to
a UHPLC instrument in line with a mass spectrometer and the products eluted using a 0.1% formic
acid in water / acetonitrile gradient. Fractions containing peaks of interest were analyzed by
UHPLC-UV-HRMS plus tandem mass spectrometry (MS/MS). The isolated samples were
characterized by multidimensional NMR spectroscopy (COSY, HSQC and HMBC) using a 1.7
4
7
mm MicroCryoProbe and quantified by external calibration against a benzoic acid standard
solution applying ERETIC2.48 As a result, descarboethoxy alcohols 2 (7.98 g, 24.0 nmol) and 3
(2.22 g, 6.8 nmol) were generated where modification occurred on the A- and C-rings of
loratadine (Table 1). It is worth mentioning the isolable nanomole scale quantities of 2 and 3 are
typical and manageable amounts for our workflow which permits rapid synthesis and
pharmacological testing on miniaturized scale using nominal amounts of staring material (e.g. 306
g of 1), reagents and solvent. Descarboethoxy alcohols 2 and 3 are known human metabolites of
4
9
loratadine. Compound 2 was found to be a potent inverse agonist of human histamine H receptor
1
ACS Paragon Plus Environment
2
4

Page 25 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with a dissociation constant of 48 nM, effectively 4-fold more active then loratadine (K = 190
d
nM). Remarkably, the regioisomeric alcohol 3 (K = 28 nM) is 7-fold more potent then loratadine.
d
In addition, compounds 2 and 3 show slightly improved human hepatocyte stability compared to
loratadine (1.75 and 3.14 L/min/million cells vs. 6.1 L/min/million cells), possibly the result of
decreased lipophilicity reflected by the lower SFlogD values.50 Interestingly, we also observed
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-6
good passive permeability of 2 and 3 (RRCK Papp = 79.9 and 78.4 x 10 cm/sec, respectively). It
is important to recognize that we did not determine absolute stereochemical configuration of the
hydroxyl group for 2 and 3 and for all subsequent loratadine analogs generated herein.
Nevertheless, from an SAR chemical space viewpoint analogs 2 and 3 indicate a basic amine is
3
tolerated on the A-ring of loratadine and polarity is accepted at the C(sp )–H sites of the C-ring.
Table 1. SAR of Loratadine Analogs Generated from Rabbit Liver Microsomes.
R
N
A
N
B
D
C
Cl
R1 R2
Kd
HLM
/ RRCK SFlogD
Compound
Loratadine
R
R1
R2
HHEP
O
H
H
H
190
48
137 / 6.1
11.2
4.6
OEt
(1)
2
H
OH
ND / <1.75 79.9
0.94
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
H
H
OH
28
ND / 3.14
78.4
1.1


Human histamine H receptor dissociation constant (nM). Geometric mean of three replicates.
1
Human liver microsome intrinsic clearance CL (L/min/mg).
int
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0

Human hepatocyte intrinsic clearance CL (L/min/million cells).
int
Apparent passive permeability in the Ralph Russ Canine Kidney cell line Papp (10 cm/sec).
-6
Shake-flask logD.
ND = Not determined.
We next focused on isolating a diverse pool of A- and C-ring analogs generated from our
microbial screen. The diversity map shows microorganisms Cunninghamella elegans ATCC
8
688a (well 73), Streptomyces violascens ATCC 31560 (well 27) and Streptomyces setae ATCC
3774 (well 28) are proficient at producing products (m/z 311, M-72), (m/z 399a, M+16), (m/z
3
399b, M+16) and (m/z 415c, M+32). Therefore, we initiated product scale-ups with these three
microorganisms, which involved inoculating cultures from the previously mentioned microbes.
After two days of culture growth, the incubate was charged with 1 (5 mg, 13 mol) and the
incubation was continued. The reaction mixtures were quenched and subjected to purification and
characterization. Consequently, microorganism Cunninghamella elegans ATCC 8688a gave 4
(
574 g, 1847 nmol) as the A-ring altered descarboethoxy analog (Table 2). Compound 4 is
recognized as a known metabolite of loratadine23,51 and an approved second generation tricyclic
5
2
antihistamine drug identified as desloratadine, which displays potent affinity for the H receptor
1
(
K = 24 nM). The C-ring positional hydroxyl-isomers 5 (27 g, 67 nmol) and 6 (11 g, 27 nmol)
d
were generated from Streptomyces violascens ATCC 31560 and are not reported in the literature
which speaks to the ability of our approach to deliver novel chemical matter for drug discovery.
Hydroxyl analog 5 (K = 454 nM) was less active then its regioisomer 6 (K = 257 nM) which
d
d
ACS Paragon Plus Environment
2
6

Page 27 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
maintained H potency. Compound 6 also shows 2-fold improved permeability RRCK P = 26
1
app
-6
-6
x 10 cm/sec compared to 1 (RRCK P = 11 x 10 cm/sec). Based on the SAR results of 5 and
app
6, it appears C-ring modification is better tolerated without the A-ring carboethoxy group as
previously shown with congeners 2 and 3 which illustrates the receptor’s high degree of selectivity.
Lastly, microorganism Streptomyces setae ATCC 33774 afforded A-ring modified
dicarbinolamide 7 (43 g, 103 nmol), which is an unreported analog of loratadine. The stability
of 7 was examined at several pH values (e.g. pH 1.2, 7.4 and 10.0) and the compound was found
to be unchanged over 24 hours by HPLC-MS (Figure S2 in the Supporting Information). The
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
3
dicarbinolamide functional group can be chemically prepared via a Hofmann rearrangement. To
our delight, compound 7 was found to have favorable potency (K = 51 nM) and better metabolic
d
stability in human liver microsomes (HLM CL = 21 L/min/mg) in comparison to loratadine
int
(HLM CL = 137 L/min/mg). In addition, the human hepatocyte values for 7 (HHEP CL = 13
int
int
L/min/million cells) and 1 (6 L/min/million cells) are similar and reflect good cellular metabolic
stability. Importantly, loratadine analogs 5, 6 and 7 are new compounds and underscore the
capability of our approach to access novel chemical space coupled with new SAR vectors for
medicinal chemistry teams to explore from a single transformational step.
Table 2. SAR of Loratadine Analogs Generated from Microorganisms.
R
R3
N
A
R4
N
D
B
C
Cl
R1 R2
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 88
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Kd
HLM / RRCK SFlogD
HHEP
Compound
R
H
R1
R2
R3
R4
4
5
H
H
H
H
H
H
H
24
4.8 / 6.5
2.8
1.3
3.3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
O
O
OH
454
132 / 61.5 23.7
OEt
OEt
OEt
6 
OH
H
H
H
H
H
257
157 / 40
21 / 14
26.2
19.8
3.2
2.3
7
OH OH 51


Human histamine H receptor dissociation constant (nM). Geometric mean of three replicates.
1
Human liver microsome intrinsic clearance CL (L/min/mg).
int

Human hepatocyte intrinsic clearance CL (L/min/million cells).
int
Apparent passive permeability in the Ralph Russ Canine Kidney cell line Papp (10 cm/sec).
-6
Shake-flask logD.
From Cunninghamella elegans ATCC 8688a.
From Streptomyces violascens ATCC 31560.
From Streptomyces setae ATCC 33774.
We applied our protein-ligand docking protocols of known CYP isoforms (e.g. CYP1A2,
2
C8, 2C9, 2C19, 2D6 and 3A4) to analog 7 with the aim to understand the nature of the improved
HLM metabolic stability compared to loratadine. Evaluating the CYP2D6 model with analog 7
as shown in Figure 12, we observe the oxygen atom of the iron(IV)-oxo porphyrin radical cation
to be 13 Å from the methylene hydrogen atom of the ethyl carbamate group in 7 placing this moiety
ACS Paragon Plus Environment
2
8

Page 29 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of the molecule far from the CYP active site which could explain the lower intrinsic clearance
value.54 From a loratadine drug design perspective, the protein-ligand model suggests that
introducing polar functionality adjacent to the carbamate nitrogen atom on the A-ring can improve
metabolic stability. The diversity map shows compound 7 was not detected by HPLC-MS in the
BM3 or liver microsomes and recombinant CYP screens.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CYP2D6 Model with Analog 7
Analog 7

1
3 A
Iron(IV)-oxo
porphyrin
Figure 12. Modeled pose of analog 7 with CYP2D6. The oxygen atom of the CYP2D6 iron(IV)-
oxo porphyrin radical cation is 13 Å from one of the methylene hydrogen atoms of the ethyl
carbamate moiety in 7.
With access to analog 5, we attempted to explore the C-ring vector and restore activity by
synthesizing the fluorinated congener via a deoxyfluorination reaction using the reagent N,N-
diethylaminosulfur trifluoride (DAST).55 We have found deoxyfluorination of hydroxylated
compounds an effective strategy for drug design to optimize ligand-target potency, selectivity and
ACS Paragon Plus Environment
2
9