Enantiomeric radiochemical synthesis of R and S (1-(6-amino-9H-purin-9-yl)- 3-fluoropropan-2-yloxy)methylphosphonic acid (FPMPA)

Article abstract of DOI:10.1002/jlcr.1505

Therapy for human immunodeficiency virus (HIV)-infected patients requires chronic multidrug administration. The eventual failure of therapy in some patients has brought into question the tissue concentration of the drugs. With an appropriately radiolabele

Full text of DOI:10.1002/jlcr.1505

Research Article
Received 28 November 2007,
Revised 31 January 2008,
Accepted 31 January 2008
Published online 13 March 2008 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1505
Enantiomeric radiochemical synthesis of R
and S (1-(6-amino-9H-purin-9-yl)-3-
fluoropropan-2-yloxy)methylphosphonic acid
(FPMPA)
Ã
Dale O. Kiesewetter,a Kathleen Knudson,a Matt Collins,b
Sharat Srinivasula,c Esther Lim,d and Michele Di Mascioe
Therapy for human immunodeficiency virus (HIV)-infected patients requires chronic multidrug administration. The eventual
failure of therapy in some patients has brought into question the tissue concentration of the drugs. With an appropriately
radiolabeled compound, we could utilize positron emission tomography to provide quantitative time–activity curves for
various tissues. We have developed a fluorine-18 labeled analog of Tenofovir, the active metabolite of Tenofovir DF, a
commonly prescribed component of multidrug therapy. Because (1-(6-amino-9H-purin-9-yl)-3-fluoropropan-2-yloxy)methyl-
phosphonic acid (FPMPA) has a chiral center, we prepared both enantiomers and confirmed that the S-isomer exhibited
significantly higher antiviral activity than the R-isomer. In viral replication inhibition assays in human MT4 cells infected
with SHIVDH12R, S-FPMPA had an IC50 of 1.85 lM (95% CI; 0.8–5.53), while the R-isomer was inactive. An appropriate chiral
precursor was prepared to allow the incorporation of fluorine-18. The [18F]FPMPA in racemic, R, or S form was prepared in a
50 min synthesis in 3875% yield (n = 23, corrected for decay). The product was of high radiochemical and enantiomeric
purity. The specific activity of the final product was 4.071.8 Ci/lmol at EOB (end of bombardment). This product may
provide information about drug tissue distribution in animal models under chronic drug treatment.
Keywords: fluorine-18; Tenofovir; antiretroviral drugs; HAART
concentrations in tissues. However, due to the invasive nature of
these procedures, it is difficult, for obvious ethical reasons, to
Introduction
Since the introduction of highly active antiretroviral therapy
(HAART), there has been a dramatic decrease in human
immunodeficiency virus type 1 (HIV-1)-related mortality, and
HIV-1 infection has been transformed into a chronic or subacute
disease in substantial numbers of treated patients1. HAART
treatment successfully suppresses plasma viremia in many
patients below the threshold of detection of the commonly
used viral determination assays in clinical practice (25 virions per
mL of plasma). However, more sensitive assays have demon-
strated that CD41 T cells, the major cell target for HIV-1, retain
replication-competent viral DNA and may be reactivated to
produce virus even after years of viral suppression2. Thus, long-
term anti-HIV treatment does not permit the eradication of HIV-
1 infection and continues to be associated with substantial
toxicity3, adherence difficulties, and drug resistance evolution4.
Our current knowledge on antiviral efficacy, dosing, and toxicity
of available HAART regimens are mostly derived from plasma or
blood kinetics of anti-HIV drugs. However, the blood comprises
only 2% of total target cells in the body. Tissue drug levels may
substantially differ from corresponding plasma levels, and drug
distribution processes may be characterized by high intertissue
variability, leading to suboptimal target site concentrations and
the potential risk for therapeutic failures5. Biopsies, for certain
lymphoid organs, have frequently been applied to measure drug
perform longitudinal analysis in the same subject, which sets a
limit on our knowledge of the in vivo kinetics of antiretroviral
drugs during the years of chronic treatment in the population of
HIV-1-infected patients. One component often included in
HAART regimens is Tenofovir disoproxil fumarate (Figure 1),
which is rapidly converted in vivo to Tenofovir (PMPA {1-(6-
amino-9H-purin-9-yl)- propan-2-yloxy)methylphosphonic acid}).
This drug is a member of a class of nucleotide analog reverse
transcriptase (RT) inhibitors characterized by low metabolism in
aPositron Emission Tomography Radiochemistry Group, National Institute of
Biomedical Imaging and Bioengineering, NIH, Bethesda, MD, USA
bBioqual, Rockville, MD, USA
cBiostatistics Research Branch, SAIC-Frederick, Inc., NCI-Frederick, Frederick, MD
21702, USA
dPhiladelphia VA Medical Center, University of Pennsylvania, Philadelphia, PA,
USA
eBiostatistics Research Branch, National Institute of Allergy and Infectious
Diseases, NIH, Bethesda, MD, USA
*Correspondence to: Dale O. Kiesewetter, PRG/NIBIB, 10 Center Drive MSC 1180,
10/1C401, Bethesda, MD 20892, USA.
E-mail: dk7k@nih.gov
J. Label Compd. Radiopharm 2008, 51 187–194
Copyright r 2008 John Wiley & Sons, Ltd.

Dale O. Kiesewetter et al.
Figure 1. The structures of Tenofovir and its orally active prodrug.
Figure 2. Chemical synthesis of authentic (RS) FPMPA.
the body6. The historical development of this class of
HPMPA, which contains a 30-hydroxy moiety, was more active
compounds was recently the subject of a review article7. The against DNA viruses while PMPA, which is unsubstituted, was
molecule contains a phosphonate moiety to mimic the 50- more active against RNA viruses. FPMPA was also more active
phosphate of the normal nucleotide. Thus, it is converted against RNA viruses10. These molecules possess a chiral center at
intracellularly, by infected and uninfected cells, into Tenofovir the second position of the propyl chain. The S enantiomer of
diphosphate, an acyclic analog of ATP that inhibits viral RT8. FPMPA and R enantiomer of PMPA showed the highest
Positron emission tomography (PET) has greatly expanded the inhibition of retroviral replication10.
scope of the pharmacokinetic measurements that can be
In this article, we report the inhibition of viral replication of
performed noninvasively in humans. Thus, we sought to the two isomers in rhesus monkey peripheral blood mono-
identify an analog of Tenofovir to be labeled with a positron nuclear cells (PBMC) and human MT4 cells. We also report the
emitting radionuclide as a candidate PET agent to study successful radiochemical synthesis of racemic [18F]FPMPA as well
antiretroviral tissue kinetics in vivo with noninvasive imaging.
The chemical structure of PMPA precludes the rapid
incorporation of carbon-11 (t 1/2 20 min) into the molecule.
There has been a significant amount of work in the develop-
as of the pure enantiomers (R)- and (S)-[18F]FPMPA.
Results and discussion
ment of nucleotide analog RT inhibitors. Fortunately,
a
The ability of antiretroviral compounds to penetrate into tissue
fluorinated analog (1-(6-amino-94-purin-9-yl)-3-fluoropropan-2- target cells at sufficient levels to prevent residual viral
yloxy (FPMPA), Figure 2, compound 4) had already been replication is crucial for the long-term control of viremia in
prepared9. The hydroxyl analog (HPMPA) is also a known HIV-1-treated patients. It is known that antiretroviral compounds
antiviral compound. The prior synthesis of HPMPA provided the have differential penetrations in certain anatomic compart-
necessary knowledge for the preparation of a suitable precursor ments11,12 and between subjects13, raising the possibility that
for radiofluorination.
localized sub-optimal concentrations of one drug of the
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 187–194

Dale O. Kiesewetter et al.
combination increases the probability of drug resistance strain The yield of 7 was greatly enhanced by conducting the reaction
generation14, a major cause of drug therapy failure in the in pyridine.
population of HIV-1-infected patients. In order to be effective
Preparation of the pure enantiomers of the methanesulfonate
these drugs need to penetrate within cells, the intracellular was achieved using identical chemistry. The enantiomeric
concentration is the ultimate driving force to define antiviral diethylphosphonates of HPMPA were prepared by first reacting
efficacy. The tissue and intracellular accumulation of HIV drugs is adenine with the commercially available R or S glycidyl butyrate
determined primarily by their physicochemical characteristics in DMF in an analogous procedure to that previously
(e.g. lipophilicity, charge), by plasma protein binding and active described23. The enantiomeric purity of the intermediates
transporters, such as P-glycoprotein, and the more recently was not determined. However, as none of the synthetic steps
discovered family of multidrug resistance proteins15. It is involve cleavage of chemical bonds at the chiral center,
currently unknown whether differential penetration of drugs in we expect the chiral purity to be unchanged from
certain anatomic compartments is associated with long-term that of the commercial precursors. The butyrates were
control of viremia in HIV-1-infected treated patients, mostly methanolized to provide the single enantiomer of the diol.
because of the lack of techniques that can noninvasively This diol was converted to the bis(trityl) compound, 6, as a
interrogate local drug-level concentrations throughout the single enantiomer. The rest of the procedure follows that
body. Indeed, our current knowledge on antiretroviral drug described above.
kinetics comes from studies mostly in blood compartments,
This methanesulfonate, 7, as a racemic mixture or as
along with a few studies in cerebrospinal fluid, seminal, individual R or S enantiomers provided for efficient fluorine-18
and extra-vaginal fluid kinetics from which inference is made incorporation under conditions employing Kryptofix 2.2.2 and
on the central nervous system and genital tract compartments. K2CO3 in acetonitrile (Figure 3). Protection of the 2-amino group
It is also unknown whether, in tissues, changes in active of adenine was not required. The deprotection of the diethyl
efflux pumps due to prolonged exposure to antiretroviral phosphonate proved to be rapid using trimethylsilyl bromide,
therapy may explain the waning HAART efficacy during chronic requiring only 5 min (Figure 4). Purification was accomplished by
antiviral therapy, thus, mimicking known paradigms of high-performance liquid chromatography (HPLC), which pro-
drug failure in anticancer and antibacterial chemotherapy16,17
.
vided the desired radiolabeled [18F]FPMPA. The radiochemical
We here describe a radiolabeled derivative of a commonly synthesis and purification were accomplished in 3875% yield
used antiretroviral compound, PMPA, as a model to noninva- (n = 23, corrected for decay) with a synthesis time of about
sively study biodistribution with PET imaging. PMPA itself 50 min. Thus, from about 30 mCi of [18F]fluoride one can obtain
was not readily amenable to radiolabeling, but a fluorine- 6–8 mCi of [18F]FPMPA in less than 1 h. The product was of
containing analog, FPMPA, was already known that displayed high radiochemical purity (498%) (Figure 5). The specific
antiviral activity10,18,19. Furthermore, of the two enantiomers, S- activity of the product was measured for eight batches and
FPMPA displayed greater inhibitor activity in the viral systems was 4.071.8 Ci/mmol at EOB. The HPLC eluate (2% ethanol in
studied10.
50 mM NaH2PO4) may only require adjustment for isotonicity
We prepared racemic FPMPA19 using the literature procedure and pH prior to injection for in vivo studies. Identity was
as a guideline (Figure 2). Adenine was treated with epifluor- confirmed by co-elution of the obtained radiochemical product
ohydrin to provide 3-fluoro-2-hydroxypropyl adenine. The with the unlabeled standard on HPLC. Chiral HPLC
phosphonate functionality was introduced by coupling (diethox- of the products demonstrated a single enantiomer in each case
yphosphoryl)methyl trifluoromethanesulfonate (2)20 with the (Figure 6).
secondary alcohol of 1 without protection of the 2-amino
functional group of adenine. Treatment of 3 with bromotri-
methylsilane yielded racemic FPMPA. The synthesis of the
respective enantiomers of authentic FPMPA was achieved
Inhibitory activity of R-PMPA, (RS)-FPMPA,
R-FPMPA, and S-FPMPA
ultimately by preparation of chiral methanesulfonate precursors We conducted viral replication inhibition tests in two systems,
(described below) and conducting a fluorination.
rhesus monkey PBMC infected with SIVmac239 and human
For purposes of incorporating fluorine-18 (t 1/2 = 110 min), MT4 cells infected with SHIVDH12R, which are of interest for our
we required a precursor activated for nucleophilic fluorination. subsequent work. Our assay of R-PMPA inhibition of SIVmac239
Following fluorination any and all protective groups must provided an IC50 of 2.32 mM (95% CI; 1.51–3.34] that
be easily removable to allow total synthesis time to be was consistent with previously reported data24 (Figure 7 (a)).
minimized as required by the short half-life of the radionuclide. Two separate assays with (RS)-FPMPA provided viral
We prepared the bis(trityl) diethylphosphonate analog of growth IC50s of 4.85 mM (95% CI; 3.8–6.67) and 6.66 mM (95%
HPMPA (5) in three steps from adenine. Following synthesis CI; 5.1–8.9) (Figure
7 (b)), which were not statistically
of the 2, 3-dihydroxypropyl adenine, two trityl groups significantly different. However, the observed increase in IC50
were introduced at the primary alcohol and the amine using for the racemic (RS)-FPMPA of approximately 2-3-fold compared
the procedures described by Webb21 or Jokic and Skaric22. with the IC50 of R-PMPA was highly statistically significant
We obtained slightly better yields of the bis(trityl) intermediate (Po0.001) and consistent with the racemic antiretroviral drug
using the procedure conducted in pyridine22. The phosphonate being a 1:1 ratio of two enantiomers, of which only one is active
functionality was introduced using the same procedure as for on the RT10. Thereafter, the two isolated enantiomers were
FPMPA. Removal of the trityl protecting groups with acetic acid subsequently tested for viral replication in human MT4 cells
provided the diethyl phosphonate 6. Treatment of 6 with infected with SHIVDH12R. As expected the S-FPMPA, but not the
methanesulfonyl chloride in dimethyl formamide (DMF) R-FPMPA, showed inhibitory activity of viral replication with an
and triethyl amine gave a small amount of a product in which estimated IC50 of 1.85 mM (95% CI; 0.8–5.53) as shown in Figure
the methanesulfonate had formed at only the primary alcohol. 7(c) and (d).
J. Label Compd. Radiopharm 2008, 51 187–194
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

Dale O. Kiesewetter et al.
Figure 3. Chemical synthesis of methanesulfonate precursor for radiolabeling. Compound 5 can be prepared in each enantiomeric form.
Figure 4. Radiochemical synthesis scheme.
1-(6-Amino-9H-purin-9-yl)-3-fluoropropan-2-ol
(1)
was
Materials and methods
prepared by the reaction of adenine with epifluorohydrin
according to the published procedure19. Instead of crystal-
lization, we used silica gel chromatography to isolate the
title compound.
General: Unless otherwise indicated all chemicals were pur-
chased from commercial sources and used as received.
(Diethoxyphosphoryl)methyl trifluoromethanesulfonate (2) was
prepared by Xu et al.’s20 method. 1H-NMR spectra were obtained
on a Varian Gemini 2000 spectrometer at 200 MHz or a Bruker
Avance 300 at 300 MHz. 13C-NMR were obtained at 75 MHz on
the Bruker Avance 300. Signals are reported downfield from
Diethyl (1-(6-amino-9H-purin-9-yl)-3-fluoropropan-2-yloxy)-
methylphosphonate (3)
internal tetramethylsilane. Semipreparative HPLC employed a 1 (292 mg, 1.39 mmol) was dissolved in DMF (5 mL) and treated
Perkin-Elmer Series 200 pump coupled with an HP1100 detector with NaH (110 mg, 60% emulsion, 2.76 mmol) and the mixture
and data system. Analytical HPLC used an HP1100 system.
stirred at room temperature for 80 min. 2 (470 mg, 1.57 mmol)
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd. J. Label Compd. Radiopharm 2008, 51 187–194

Dale O. Kiesewetter et al.
Figure 5. Sample quality control chromatogram of S-FPMPA. Column: Agilent Eclipse XDB-C-18 4.6 ꢀ 150 mm. Eluate: 15% CH3CN, 85% 5 mM nBu4NH2PO4, 1 mL/min.
Figure 6. Radiochromatograms of the isolated enantiomers. Column: Luna C-18 4.6 ꢀ 250 mm. Eluate: 4 mM CuSO4, 4 mM L-phenyl alanine, 1 mL/min.
was added dropwise over about 5 min and stirred at room
temperature for 3 h. At the end of the reaction, the DMF was
removed on a rotary evaporator (15 T). Toluene (40 mL) was
(1-(6-Amino-9H-purin-9-yl)-3-fluoropropan-2-yloxy)methyl-
phosphonic acid (4)
added and evaporated on the rotary evaporator. The resulting 3 (204 mg, 0.565 mmol) was dissolved in 3 mL DMF and
slurry was preabsorbed on silica gel for flash chromatography. placed under argon. Bromotrimethylsilane (722 mL, 5.65 mmol)
Chromatography was conducted on an Analogix system using was added via a syringe. The reaction was stirred overnight. The
an Analogix RS-12 cartridge and eluting with 90% chloroform, DMF was evaporated on a rotary evaporator; toluene was added
9% methanol, 1% ammonium hydroxide. The product was and evaporated. The residue was dissolved in 3 mL water; 3 mL
collected (252 mg, 50.2%). 1H-NMR (200 MHz, CDCl3) d 1.27 (t, acetone was added and the resulting solution cooled. The
J = 7.2 Hz, 3H), 1.32 (t, J = 7.4 Hz, 3H), 3.7–3.9 (m, 2H), 4.0–4.2 (m, precipitate was collected and recrystallized from ethanol/ether
4H), 4.2–4.2 (m, 2H), 4.4–4.65 (m, 2H), 4.65–4.81 (dd, J = 10, to give the product (112 mg, 65%). 1H-NMR (200 MHz, DMSO-d6)
3.4 Hz, 1H), 5.59 (brs, 2H), 7.94 (s, 1H), 8.36 (s, 1H).
d 3.50–3.75 (m, 2H), 3.95–4.20 (dm, J = 21.5, 1H), 4.20–4.60 (m,
J. Label Compd. Radiopharm 2008, 51 187–194
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

Dale O. Kiesewetter et al.
Figure 7. Inhibitory activity of R-PMPA (a) and RS-FPMPA (b) in macaque PBMC cells infected with SIVmac239; and of S-FPMPA (c) and R-FPMPA (d) in human MT4 cells
infected with SHIVDH12R
.
3.5 H), 4.60–4.75 (dd, J = 10.2, 3.4 Hz, 0.5H), 7.29 (s, 2H), 8.14 (s, 61.5, 61.5, 61.6, 61.7, 63.6, 80.0 (d, J = 11.4 Hz), 118.3, 141.3,
1H), 8.15 (s, 1H). 19F-NMR (internal TFA) ꢁ232.5 (dt, J = 47.2, 149.6, 152.2, 155.8.
18.4 Hz). LCMS (ESI), 306 M1H.
(R)-3-(6-Amino-9H-purin-9-yl)-2-((diethoxyphosphoryl)-
methoxy)propyl methanesulfonate (7)
(R)-Diethyl (1-(6-(tritylamino)-9H-purin-9-yl)-3-(trityloxy)-
propan-2-yloxy)methylphosphonate (5)
Alcohol 6 (500 mg, 1.39 mmol) was dissolved in 10 mL pyridine
(R)-1-(6-(Tritylamino)-9H-purin-9-yl)-3-(trityloxy)propan-2-ol (2.28 g, and treated with methanesulfonyl chloride (191 mg, 1.67 mmol).
3.29 mmol) in DMF (15 mL) was treated with NaH (263 mg The reaction was followed by thin layer chromatography (TLC).
(60% in oil)) under argon for 2 h. The mixture was placed in an After 1 h, an additional equivalent of methanesulfonyl chloride
ice bath and 2 was added dropwise. The solution was stirred was added, followed an hour later by a third equivalent. The
at ice temperature for 1 h then set in a refrigerator over- pyridine was evaporated; the residue dissolved in ethanol and
night. The DMF was evaporated under vacuum and the residue preabsorbed onto 3.5 g of silica gel. The material was
absorbed onto 6 g of silica gel. The product was purified chromatographed under the following program: 100% ethyl
by chromatography eluting with 1:1 ethyl acetate:hexane acetate for 1 min followed by a gradient over 20 min to 100%
for 7 min then over 1 min transitioning to 100% ethyl acetate. (20% (10% NH4OH in ethanol) 80% EtOAc). The yield was
Yield 1.89 g, 68%. The S-isomer was prepared with the 455 mg, 75 %. The S-isomer was similarly obtained in 88% yield.
same procedure starting from the S precursor in 78% yield. 1H-NMR for R isomer (300 MHz, DMSO d-6) d 1.13 (t, J = 7.1 Hz,
1H-NMR for R isomer (300 MHz, acetone d-6) d 1.17 (t, J = 7.1 Hz, 3H), 1.18 (t, J = 7.1 Hz, 3H), 3.34, (s, 3H), 3.8–4.1 (m, 6H), 4.2–4.6
3H), 1.20 (t, J = 7.4 Hz, 3H), 3.10 (dd, J = 10.5, 5.0 Hz, 1H), 3.28 (dd, (m, 5H), 7.24 (s, 2H), 8.10 (s, 1H), 8.16 (s, 1H). 13C for S-isomer
J = 10.4, 4.2 Hz, 1H), 3.7–3.9 (m, 1H), 3.9–4.15 (m, 6H), 4.3–4.6 (m, (75 MHz, DMSO d-6) d 16.1 (t, J = 5.4 Hz), 36.7, 42.9, 61.6 (d,
2H), 6.97 (s, 1H), 7.1–7.3 (m, 18H), 7.3–7.5 (m, 12H), 7.82 (s, 1H), J = 6.4 Hz), 61.6, 61.8 (d, J = 6.4 Hz), 77.1 (d, J = 13.1 Hz), 118.5,
8.02 (s, 1H).
149.6, 152.4, 155.9.
R- or S- (1-(6-Amino-9H-purin-9-yl)-3-fluoropropan-2-yloxy)-
methylphosphonic acid (FPMPA, 9)
(R)-Diethyl (1-(6-amino-9H-purin-9-yl)-3-hydroxypropan-2-
yloxy)methylphosphonate (6)
The corresponding chiral methanesulfonates 7 were suspended
in CH3CN and treated with 3-fold excess of tetramethylammo-
nium bifluoride and heated at 801C for 3 h until TLC indicated
consumption of starting material. The product was loaded
onto a silica gel column and eluted with 90:9:1 (CHCl3:MeOH:
NH4OH). The major component was dissolved in 0.5 mL DMF
and treated with 10 equivalents of bromotrimethylsilane over-
night. The DMF and bromotrimethylsilane were removed
under vacuum, the residue taken up in water and the product
precipitated with acetone. The products were about 85% pure
by HPLC (Agilent XDB C-18, 85% (5 mM tetrabutylammonium
phosphate, TBAP) 15% CH3CN). Higher purity (98%) was
obtained after preparative HPLC (Luna C-18 2% ethanol in
Intermediate 5 (1.89 g, 2.24 mmol) was treated with 50 mL of
80% HOAc in water. The mixture was heated at 801C for 2 h. The
mixture was allowed to stand at room temperature for 2 h then
placed in a refrigerator overnight. The mixture was filtered to
collect some of the triphenylmethanol side product. The filtrate
was evaporated and purified by chromatography. The sample
was preabsorbed on 3 g silica gel and eluted with 90:9:1
(CHCl3:MeOH:NH4OH) to give the product in 66% yield. The
S-isomer was similarly obtained in 69% yield 1H-NMR for R-
isomer d 1.11 (t, J = 7.0 Hz, 3H), 1.16, (t, J = 7.0 Hz, 3H), 3.47 (t,
J = 5.3 Hz, 2H), 3.7–3.9 (m, 7H), 4.20 (dd, J = 14.5, 7.4 Hz, 1H), 4.35
(dd, J = 14.5, 3.6 Hz, 1H), 4.98 (t, J = 5.7, 1H, OH), 7.20 (s, 2H), 8.04
(s, 1H), 8.13 (s, 1H). 13C-NMR for S-isomer, 16.1, 43.4, 59.9, 61.4,
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 187–194

Dale O. Kiesewetter et al.
50 mM NH4OAc) of part of the sample. NMR indicated that Each assay was performed in triplicate. On day 7, p24 ELISA
the samples contained 1–2 equivalents of acetate after drying assay was performed on culture supernatant to determine viral
under vacuum.
burden30. For both assays, results were expressed as
a
1H-NMR for R isomer (300 MHz, D2O) d 3.51 (dd, J = 13.1, percentage of growth of the virus.
9.7 Hz, 1H), 3.69 (dd, J = 13.1, 9.2 Hz, 1H), 4.04 (dm, J = 22 Hz, 1H),
4.3–4.60 (m, 3.5H), 4.71 (dd, J = 10.6, 3.6 Hz, 0.5 H—part of CH2F).
R-, S- or RS (1-(6-Amino-9H-purin-9-yl)-3-[18F]fluoropropan-2-
Statistical methods
yloxy)methylphosphonic acid ([18F]FPMPA, 9). A reaction tube
(13 ꢀ 100 mm) was charged with Kryptofix 2.2.2 (4.5 mg,
12 mmol), K2CO3 (60 mL of 0.1 M in water, 6 mmol), and F-18
fluoride. The solution evaporated to dryness. Three portions of
CH3CN were added and each in turn evaporated to dryness. A
solution of methanesulfonate 7 (3 mg, 8 mmol) in 300 mL of
CH3CN was added to the dried residue. The resulting mixture
was heated at 1051C for 5 min. The solution was loaded onto a
short column of silica gel. The reaction tube was rinsed with
200 mL of 15% ethanol in CH3CN; this solution was also added to
the silica gel column. The combined liquids were pushed
through the silica gel column and the column was eluted with
an additional 600 mL of 15% ethanol in CH3CN. The combined
elute was assayed and evaporated to dryness. CH3CN (100 mL)
and bromotrimethylsilane (100 mL) were added. The resulting
solution was heated at 751C for 1 min then allowed to stand at
room temperature for 4 additional minutes. The solution was
evaporated to dryness; the residue was taken up in 200 mL of
HPLC eluant and injected onto the HPLC (Luna C-18(2) 10 ꢀ 250,
eluted with 2% ethanol in 10 mM NaH2PO4). The radioactive
peak eluting at ꢂ15 min was collected. The eluant, after
adjusting for isotonicity and pH, may be suitable for direct
injection into animals. Specific activity was measured by re-
injection onto an Agilent XDB-C18 column, eluting with 15%
CH3CN in 5 mM TBAP and comparing the mass peak at 250 nm
with a standard curve (tR = 7.7 min).
The IC50 values are determined by nonlinear regression
analysis using the equation of a sigmoid plot. For each
assay, the mean value of the replicates at different concentra-
tions is fitted to the sigmoid curve using the Levenberg–Mar-
quardt Algorithm31.
The best fit theoretical curve is then used to locate the IC50,
the concentration of the antiviral compound needed to inhibit
50% of viral growth. The 95% confidence interval of IC50 is
calculated by a bootstrap method32. With this method, the best
fit theoretical curve is challenged by the original observed data
(mean value of the replicates) by generating surrogate data,
obtained by shuffling without replacing the residual difference
between the theoretical and the observed data and adding the
residual to the observed data. This procedure is repeated 1000
times, and the 95% bootstrap percentile interval is calculated
from these bootstrap IC50’s. The P-value of the difference
between estimates of IC50 in two independent plates was
obtained from a permutation test on the distribution of the
bootstrap values.
Conclusions
We have confirmed in our system of retroviral replication that
the S-FPMPA is the more potent enantiomer. We have
developed a robust radiochemical synthesis method for the
preparation of R- or S-[18F]FPMPA that is suitable for preclinical
studies of the utility for imaging drug biodistribution. The
radiochemical synthesis provides the desired product in 3875%
yield and requires about 50 min. This product is suitable for
further studies of biodistribution. Potential pre-clinical and
clinical applications of this novel PET tracer are (1) tissue drug
kinetics during chronic treatment; (2) drug interactions; and (3)
generalization of the concept of antiviral efficacy of HAART
regimens.
Determination of enantiomeric purity: A Phenomenex Luna C-
18 (4.6 mm ꢀ 250 mm ꢀ 5 mm) was eluted with 4 mM CuSO4 and
4 mM L-phenylalanine at 1 mL/min. UV absorbance was mon-
itored at 260 nm (retention time for S enantiomer 32 min; for R
enantiomer 24 min).
Assay of virus replication in rhesus monkey
PBMC and human MT4 cells
Acknowledgements
The inhibitory activity of nonradiolabeled (RS)-FPMPA and R-
PMPA was assessed in macaque PBMC infected with SIVmac239.
The preparation and infection of macaque PBMC has been This work was presented in part at the 17th International
previously described25,26. Briefly, PBMC were stimulated with Symposium on Radiopharmaceutical Sciences in Aachen,
concavalin A and cultured in the presence of recombinant Germany, April 30–May 4, 2007, S141. This work was supported
human IL-2 for 3 days prior to viral inoculation. Subsequently, in part by the Intramural Research Programs of the National
PBMC were cultured in 96 well plates in the presence of Institute of Biomedical Imaging and Bioengineering.
different concentrations of (RS)-FPMPA or R-PMPA and spino-
culated (1200 ꢀ g for 1 h)27 with virus stocks at a multiplicity of funds from the National Cancer Institute, National Institutes of
infection (MOI) of 10ꢁ2
Each assay was performed in Health, under contract N01-CO-12400. The content of this
This project has been funded in whole or in part with federal
.
six replicates. On day 5 from viral inoculation, culture super- publication does not necessarily reflect the views or policies of
natants were collected from PBMC to assess virus replication the Department of Health and Human Services, nor does the
by virion-associated RT assay28. The inhibitory activity of R- mention of trade names, commercial products, or organizations
FPMPA and S-FPMPA was determined in human T- leukemic imply endorsement by the US Government. This research was
cells (MT4)29 infected with SHIVDH12R. Briefly, MT4 cells were supported in part by the National Institute of Allergy and
cultured in RPMI 1640 enriched with 10% fetal bovine serum Infectious Diseases.
and infected with SHIVDH12R with virus stocks at an MOI of 10ꢁ3
.
We acknowledge the Positron Emission Tomography Depart-
After 2 h incubation, the cells were washed to remove unbound ment of the Clinical Center for Radioisotope preparation. We
virions. The infected cells were cultured for 7 days in the also thank Dr. Tatsu Igarashi for helpful discussions on design
presence of different concentrations of R-FPMPA or S-FPMPA. and interpretation of viral replication assays.
J. Label Compd. Radiopharm 2008, 51 187–194
Copyright r 2008 John Wiley & Sons, Ltd.
www.jlcr.org

Dale O. Kiesewetter et al.
References
[17] M. M. Gottesman, I. Pastan, Annu. Rev. Biochem. 1993, 62,
385–427.
[18] J. Balzarini, A. Holy, J. Jindrich, H. Dvorakova, Z. Hao, R. Snoeck,
P. Herdewijn, D. G. Johns, E. Declercq, Proc. Natl. Acad. Sci. USA
1991, 88, 4961–4965.
[19] J. Jindrich, H. Dvorakova, A. Holy, Collect. Czech. Chem. Commun.
1992, 57, 1466–1482.
[1] F. J. Palella, Jr., K. M. Delaney, A. C. Moorman, M. O. Loveless,
J. Fuhrer, G. A. Satten, D. J. Aschman, S. D. Holmberg, N. Engl. J.
Med. 1998, 338, 853–860.
[2] J. K. Wong, M. Hezareh, H. F. Gunthard, D. V. Havlir, C. C. Ignacio,
C. A. Spina, D. D. Richman, Science 1997, 278, 1291–1295.
[3] A. Carr, K. Samaras, A. Thorisdottir, G. R. Kaufmann, D. J. Chisholm,
D. A. Cooper, Lancet 1999, 353, 2093–2099.
[20] Y. B. Xu, M. T. Flavin, Z. Q. Xu, J. Organic Chem. 1996, 61,
7697–7701.
[21] R. R. Webb, Nucleosides Nucleotides 1989, 8, 619–624.
[22] M. Jokic, V. Skaric, Tetrahedron Lett. 1994, 35, 2937–2940.
[23] A. Holy, Collect. Czech. Chem. Commun. 1993, 58, 649–674.
[24] C. C. Tsai, K. E. Follis, A. Sabo, T. W. Beck, R. F. Grant,
N. Bischofberger, R. E. Benveniste, R. Black, Science 1995, 270,
1197–1199.
[25] T. Igarashi, O. K. Donau, H. Imamichi, M. J. Dumaurier,
R. Sadjadpour, R. J. Plishka, A. Buckler-White, C. Buckler,
A. F. Suffredini, H. C. Lane, J. P. Moore, M. A. Martin, J. Virol.
2003, 77, 13042–13052.
´
[4] V. A. Johnson, F. Brun-Vezinet, B. Clotet, B. Conway, R. T. D’Aquila,
L. M. Demeter, D. R. Kuritzkes, D. Pillay, J. M. Schapiro, A. Telenti,
D. D. Richman, Topics HIV Med 2004, 12, 119–124.
[5] M. Muller, A. D. Pena, H. Derendorf, Antimicrob. Agents Chemother.
2004, 48, 1441–1453.
[6] Viread (tenofovir disoproxil fumarate) tablets product labeling.
Foster City C. G. S., Inc. 2003 October.
[7] E. De Clercq, Antiviral Res. 2007, 75, 1–13.
[8] B. P. Kearney, J. F. Flaherty, J. Shah, Clin. Pharmacokinet. 2004, 43,
595–612.
[26] H. Imamichi, T. Igarashi, T. Imamichi, O. K. Donau, Y. Endo, Y.
Nishimura, R. L. Willey, A. F. Suffredini, H. C. Lane, M. A. Martin,
Proc. Natl. Acad. Sci. USA 2002, 99, 13813–13818.
[27] U. O’Doherty, W. J. Swiggard, M. H. Malim, J. Virol. 2000, 74,
10074–10080.
[28] R. L. Willey, D. H. Smith, L. A. Lasky, T. S. Theodore, P. L. Earl,
B. Moss, D. J. Capon, M. A. Martin, J. Virol. 1988, 62, 139–147.
[29] S. Harada, Y. Koyanagi, N. Yamamoto, Science 1985, 229, 563–566.
[30] Y. M. Zhang, H. Imamichi, T. Imamichi, H. C. Lane, J. Falloon, M. B.
Vasudevachari, N. P. Salzman, J. Virol. 1997, 71, 6662–6670.
[31] P. E. Gill, W. Murray, M. H. Wright, Practical Optimization, Academic
Press, London, 1981, pp. 136–137.
[9] J. Jindrich, A. Holy, H. Dvorakova, Collect. Czech. Chem. Commun.
1993, 58, 1645–1667.
[10] J. Balzarini, A. Holy, J. Jindrich, L. Naesens, R. Snoeck, D. Schols,
E. Declercq, Antimicrob. Agents Chemother. 1993, 37, 332–338.
[11] S. Taylor, A. Pereira, HIV Med. 2000, 1(Suppl. 2), 18–22.
[12] S. Taylor, R. P. van Heeswijk, R. M. Hoetelmans, J. Workman,
S. M. Drake, D. J. White, D. Pillay, Aids 2000, 14, 1979–1984.
[13] S. Rodriguez-Novoa, P. Barreiro, I. Jimenez-Nacher, A. Rendon,
V. Soriano, AIDS Rev. 2005, 7, 103–112.
[14] T. B. Kepler, A. S. Perelson, Proc. Natl. Acad. Sci. USA 1998, 95,
11514–11519.
[15] A. Fridland, M. C. Connelly, B. L. Robbins, Antiviral Ther. 2000, 5,
181–185.
[16] C. F. Higgins, Annu. Rev. Cell Biol. 1992, 8, 67–113.
[32] B. Efron, R. J. Tibshinari, In The Bootstrap Estimate of Standard Error
(Ed.: D. R. Cox), Chapman & Hall, New York, 1993, pp. 45–49.
www.jlcr.org
Copyright r 2008 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2008, 51 187–194