Vendrame R., Ferreira M. M. C., Takahata Y., “STRUCTURE -ACTIVITY RELATIONSHIP (SAR) OF CONTRACEPTIVE PROGESTOGENS STUDIED WITH PATTERN RECOGNIGITION METHODS USING CALCULATED PHYSICOCHEMICAL PARAMETERS”. Düsseldorf, Germany, 27/08-01/09/2000: 13th European Symposium on Quantitative Structure-Activity Relationships: Rational Approaches to Drug Design, Book of Abstracts, P.41 (2000). Poster P41. Section: New Developments in Chemometrics.
STRUCTURE -ACTIVITY RELATIONSHIP (SAR) OF
CONTRACEPTIVE PROGESTOGENS STUDIED WITH PATTERN
RECOGNIGITION METHODS USING CALCULATED
PHYSICOCHEMICAL PARAMETERS
Rosana Vendrame, Márcia M. C. Ferreira and Yuji Takahata
Instituto de Química, Universidade Estadual de Campinas, Caixa Postal
6154,
Campinas, São Paulo, 13081-970, Brazil.
The combined oral contraceptives (COCs) have been widely used
all over the world for
more than three decades. The
COCs are generally obtained from the
combination between
ethynylestradiol (EE) and one of three new
progestogens: desogestrel (DSG), gestodene (GSD)
and norgestimate (NGM). Though the
percentage of the side effects is quite small,
there still
remains room for improvement. The aim
of the present work is to establish a structure-activity
relationship for a series of
steroids including the three new
generation progestogens using
calculated molecular properties.
The biological activities studied were
(I) oral contraceptive
activity (OCA), (II) androgenic
effect and (III) binding affinity
to Sex Hormone Binding
Globulin (SHBG). The
molecules included in this study
are: (1) progesterone (P),
(2)
norethindrone (NET), (3) NGM,
(4) levonorgestrel (LNG), (5)
DSG and (6) GSD, (7) 17-
deacetylnorgestimate (ANGM) and (8) 5a-dihidrotestosterone
(DHT). The steroids 3,
5 and 6 are
the three new progestogens.
Our strategy was to calculate
as many physicochemical
descriptors (parameters) as possible with
the semiempirical AM1 method since we do not know,
ahead of time, which properties
are closely related to the biological activity. Statistical
methods
are used to choose a set of the descriptors
that correlate best to the biological activity. The SAR
analysis was made using
the following pattern recognition
methods: Principal Component
Analysis (PCA) and
Hierarchical Clustering Analysis
(HCA). We used the
computer programs
called PIROUETTE.
The number of the molecules studied for the three activities, (I),
(II) and (III), is as small
as six. In order to investigate the reliability of
the methods for the small number of samples,
an
external data set of
nineteen acetoxyprogesterone analogues whose relative
oral progestational
activities(Clauberg assay) is known, were added to the
original OCA data. The results showed the
statistical methods that were
in use are reliable even with the small number
of samples.
Oral contraceptive activity of the progestogens
are associated mainly with the atomic
charges at the positions 17, 10 and 13 (Q17,
Q10 and Q13)
of steroid skeleton. Androgenic effect is
associated with the frontier radical and
electron densities at the positions
7 and 9(F(r)7,
F(r)9
and
F(e)9)
of steroid skeleton. Binding affinity
for SHBG is mainly related with Q17,
with molecular
hardness (h ) and the
frontier orbital densities at positions 5, 7 and 9(F(o)5,
F(o)7 and
F(o)9)
of steroid
skeleton. Based upon the resluts of the analysis,
requirements of an efficient oral contraceptive
progestogen were proposed.