Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Track A: Basic Science

Mo.A. 1088 - Mo.A. 1092

In Vitro
Compound IC90

Conc
Max.
(pM)
13
3
2

Rat
%F
84
28
21

Clearance Conk
(L/h/kg) Mx.
0.22 19
0.17
0.59 16

I.00
0.16
0.26

Conclusions: The preclinical pharmacokinetic properties of three r,,,,
tease inhibitors indicate acceptable oral bioavailability and suggest Th.t b
ing IC90 levels might be achieved in the clinic.
F.J. Schwende, Ph.D.,The Pharmacia & Upjohn Co., Inc., 301 Henrinetta I
Kalamazoo, MI 4900 I, USA. E-mail: [email protected].
Mo.A. 1088
THE ROLE OF CHIRAL CHROMATOGRAPHY IN THE DISCOVERY
NON-PEPTIDIC, HIV PROTEASE INHIBITOR DEVELOPMENT CAN[
Seest, Eric P, Dolak L, CastleT Chemical & Biological Screening, Uipjohn
Pharmacia & Upjohn Inc., Kalamazoo, MI 49001, USA
Objectives: We initiated the work described to speed the lead find rg p
eration of orally active, non-peptidic HIV protease inhibitors.We supp ied
with pure stereoisomers and devised analytical methods for the moriti irn
purity of the candidate molecule in stereoselective syntheses and nlc bI og
Methods: We employed HPLC techniques with columns containin chlira
Especially useful were (R,R)Whelk-O I, Chiralcel OD, and Cyclobnind i c
electrophoresis (CE) with substituted I3-cyclodextrins provided arialvs(,
HPLC, particularly in those cases for which no HPLC method coui( Ibe
Results: This worked resulted in the direct resolution of more than eight
one or two stereocenters. Many of these were final inhibitors but ot!,ers
ates that were converted to the test compounds.The latter approaci all
of many enantiomerically enriched inhibitors after a resolution canj p,-rrn.
tiomers belonging to each series of compounds could be tested iinir
pharmacological tests. We separated 125 grams of an intermediate with
that led to the most promising compound. Once one enantiomer was ju
or to any of the others (e.g., its Ki = 4 pM compared to 20, 30 and 220
three enantiomers; IC50 =30- 50 nM in HIV infected H9 cells; it is orally
and dogs) development of stereoselective syntheses proceeded on a hig
work resulted in assays by HPLC and CE that supplemented NMR techr
determination of isomeric purity in synthetic schemes and prove u-.eful f
racemization in vivo.
Conclusions: Integration of preparative chiral HPLC into a lead finding e
more efficient use of medicinal chemistry resources and reduction of the
select a development candidate. Clinical trials for the selected, novel dihy
protease inhibitor will begin in the second half of 1996.
Eric P Seest, Chemical & Biological Screening, Upjohn Laboratories,, Phar
Inc., 301 Henrietta St., Kalamazoo, MI 49001. Phone: 6 I 6-384-9397. Fax:
Mo.A. 1089
SEQUENCE DETERMINATION AND CHEMICAL CHARACTERISTI(
ANTI-HIV PROTEIN, CYANOVIRIN-N, ISOLATED FROM THE CYAN
NOSTOC ELLIPSOSPORUM
Gustafson, Kirk R*, Sowder R I1**, Henderson L**, Pannel L#, Cardellia
Shoemaker R*, Boyd, M*. Laboratory of Drug Discovery Researhi andI
National Cancer-Institute;* Frederick, MD: SAIC,** Frederick, MD, NIlDD
Objective: To chemically characterize cyanovirin-N (CV-N), a novel ant IF
selected by the U.S. National Cancer Institute (NCI) for preclinical devel(
tial prophylactic virucide.
Methods: CV N was initially isolated from an aqueous cellular extract of
um Nostoc ellipsosporum using bioassay-guided fractionation. An anti--HIV
to monitor activity through a series of chromatographic separation steps
of the active constituent was achieved by C 18 reverse-phase HPL C. CV
small protein which was characterized by SDS-PAGE, electrospray,oniza
trometry (MS), amino acid analysis and N-terminal Edman degradation. S
the parent protein by endoproteinases provided a series of over lapng
that were purified and sequenced. Location of the disulfide bridges i, C,
by fast-atom-bombardment mass spectrometric analyses of peptides en
digest of the native protein.
Results: The primary amino acid sequence of CV N was unambiguously
cal methods.The sequence is novel, with no significant homology witi pr
proteins or transcription products of known nucleotide sequences. CV NF
10 I amino acid chain containing significant sequence duplication Alignme
with residues 51 101 reveals direct homology between 16 amino acids a
amino acid changes. CV-N contains four cysteines which form two intrac
(C8-C22 and C58 C73).The disulfide cross-links are required for opt imu
however CV N is remarkably resistant to organic solvents, detergent,, hi
tions or heat. 13C and I 5N-labeled CV-N has been prepared so that th
sional solution structure can be elucidated by nuclear magnetic resonenc
Conclusions: CV N represents a unique new class of HIV inhibitory prot
cyanobacteriurnm.
Kirk R. Gustafson, Ph.D., NCI-FCRDC, Bldg. 1052, Rm 12 I, Frederic, M1D
Telephone: [phone] FAX [phone]

Dog Mo.A. 1090
-F Clearance ANALYSIS OF SEQUENCE REQUIREMENTS FOR BIOLOGICAL ACTIVITY OF
(Uh/kg) RECOMBINANT CYANOVIRIN-N,A NOVELANTI-HIVVIRUCIDAL PROTEIN
MoriToshiyuki, Shoemaker R, McMahon J, Gustafson K, Boyd M. Laboratory of Drug
55 0.13 Discovery Research and Development, National Cancer Institute, Frederick, MD
0 0.26 Objective:To evaluate amino acid sequence requirements for bioactivity of recombinant
10 0.25 cyanovirin-N (r-CV-N), a novel protein currently undergoing high priority preclinical development by the U.S. National Cancer Institute as a potential prophylactic anti HIV virucide.
tidi c HIV pro Methods: Site-directed mutagenesis of a synthetic gene for r-CV-N was performed using
Iood levels exceed mutagenic oligonucleotide primers in the polymerase chain reaction or by the unique
restriction site elimination maneuver. A variety of constructs were prepared to address
HS 7256 300 108, hypotheses regarding: the importance of the two disulfide bonds contained in the natural
mrolecule, the potential for circular permutation of two homologous regions within the coding sequence; and, the potential to reduce the size of the molecule through N- or C-terminal deletions.
OF A Results: Expression in Escherichio coli of mutant versions of the r-CV-N coding sequence
DIDATE resulted in production of recombinant proteins with markedly reduced or absent anti HIV
Iaboratories, activity Both disulfide bonds within the protein were critical for activity as well as nearly the
entire 101 amino acid sequence comprising the natural CV-N, which was originally isolated
cess for a new gen- fr-om a cultured cyanobacterium. However, some anti-HIV activity was retained in the mutant
'rynthetic chemists with circular permutation of the two homologous regions within the coding sequence.
ig of enantiomeric Conclusions: Based on these initial results, it does not appear likely that major modifications
),kal fluids, or simplifications of the full natural sequence of CV-N will result in a molecule either with,1 stationary phases. superior or perhaps even comparable anti-HIV activity. However, the potent anti-HIV activiolurns, Capillary ty of fusion proteins containing the r-CV-N sequence in either N- or C- terminal locations
as an alternative to indicates that further exploitation of hybrid constructs containing the full functional gp 120 -developed. targeting sequence of CV-N may be feasible.
ty compounds with Toshiyuki Mori, Ph.D., NCI-FCRDC, Bldg. I1052, Rm I 21, Frederick, MD 2 I1702 I201 USA
s were intermedi- Telephone: [phone] FAX [phone]
lowed the synthesis
All the enan- Mo.A.1091
and in preliminary MECHANISM OF ACTION OF THE POTENT ANTI-HIV VIRUCIDAL PROTEIN
two stereocenters CYANOVIRIN-N
dged to be superi- McMahon, James B*, O'Keefe B*, Buckheit R**, Nara P*, Gustafson K*, Shoemaker R*, Boyd
pM for the other M*. Laboratory of Drug Discovery Research and Development, National Cancer Institute;*
bioavailable in rats Southern Research Institute,** Frederick, MD
h priority basis.This Objective: To investigate the mechanism of action of cyanovirin-N (CV-N), a novel anti-HIV
niques for the protein currently undergoing high priority development by the U.S. National Cancer
or determination of Institute as a potential virucide for prophylactic applications against HIV
Methods: Cell- and virus-based bioassays were used to ascertain the mechanism of action
ffort can result in of CV-N.Time of addition experiments were used initially to estimate where in the virus life
e tirne needed to cycle CV N exerted its antiviral activity A variety of antibodies specific for CV-N and viral
ydropyrone HIV antigens were used for dot- and western-blot analyses and ELISA assays. Ultrafiltration
experiments were performed to determine if viral proteins and CV-N could bind directly
macia & Upjohn, Results: Time-of-addition experiments revealed that, for maximum antiviral activity CV N
6 16 385-5225. had to be added to cells before or shortly after addition of HIV Virus pretreatment directly
with CV N followed by removal of CV-N abolished viral infectivity indicating a direct interaction of CV-N with a viral component. Ultrafiltration experiments indicated that CV-N
CS OF A NOVEL could bind directly to soluble gp120, and that the resulting complex was no longer capable
OBACTERIUM, of inhibiting infection of cells by intact HIV. Further evidence of a direct and specific interaction of CV N and the viral envelope gp I 20 was obtained by dot-blot analysis; CV-N specifiJ I" McMahon J*, cally bound to gp120 from several HIV solates as well as SIV gpl140. ELISA experiments
Development, revealed that CV N blocked recognition by several different gp 20 epitope-specific antibod)K,# Bethesda, MD ies. Western-blot analyses of HIV lysates using CV-N and specific anti-CV-N antisera estabHIV agent recently lished that CV-N bound to gplI20 but not to gp41.
opnrent as a poten- Conclusion: CV-N mediates its potent anti-HIV virucidal activity through unique and specif
ic interactions with the viral envelope glycoprotein gp I 20.
the cyanobacteri- James B. McMahon, Ph. D., NCI-FCRDC, Bldg. 1052, Rm 12 I, Frederick, MD 21702-1201
s reen was utilized USA Telephone: [phone]; FAX [phone]. Final purification
N proved to be a Mo.A. 1092
tion mass spec CONSTRUCTION AND EXPRESSION IN E. COLI OF A SYNTHETIC GENE CODING
ielective cleavage of FOR CYANOVIRIN-N,A NOVEL ANTI-HIV PROTEIN ORIGINALLY ISOLATED FROM
peptide fragments A CULTURED CYANOBACTERIUM
V N was established Shoemaker, Robert H*,Wu, L**, Smythe A*, Gulakowski R*, McMahon J*, Gustafson K*,
erated in a tryptic Boyd M*. Laboratory of Drug Discovery Research and Development, National Cancer
Institute;* SAIC,** Frederick, MD
assigned by chemi- Objective: To develop recombinant DNA methods for characterization and production of
eviously described cyanovirin-N, a novel anti HIV protein recently selected by the U.S. National Cancer
Sconsists of a single Institute for high priority preclinical development as a potential virucide for prophylaxis
nt of residues I-50 against HIV infection.
rnd 9 conservative Methods: The unique 101 residue amino acid sequence of natural cyanovirin-N (CV-N) was
hai disulfide bonds back translated to a DNA sequence of 303 base-pairs using an Escherischia coli (E. coli)
m anti-HIV activity codon preference table and supplemented with a termination codon and restriction sites to
gh salt concentra- facilitate subsequent cloning.The DNA sequence was synthesized as 13 overlapping, comre active, 3-dimen plermentary oligonucleotides and assembled to form the full coding sequence.This was then
e (IPR) methods. used as a template in the polymerase chain reaction to provide a sufficient quantity of the
ein derived from a DNA lor cloning into an expression vector
Results: Transformation of E. coli with the vector and induction of expression resulted in
l2 702I 201 USA production of a protein with potent anti-HIV activity as demonstrated by bioassay of crude
bacterial extracts as well as the affinity-purified product. Amino acid sequencing and mass
speerral analysis confirmed production of a recombinant protein (r-CV-N) indistinguishable
from the natural CV-N isolated from Nostoc ellipsosporum.
Conclusions: Biologically active CV N has been successfully produced from E. coli using
recombinant DNA technology Scale-up is feasible to provide quantities of r-CV N required
for further preclinical development.
Robert H. Shoemaker Ph.D., NCI-FCRDC, Bldg. 1052, Rm 121, Frederick, MD 21702- 1201
USA Telephone: [phone] FAX [phone]

\0,,o
O
u0
V
C
c
0
uQ
V
a
<
O
C
Q)
5)
c
0
U
C
0
o
cO
7
a)
70