Spermicidal, anti-HIV, and micellar properties of di- and trihydroxylated cationic surfactants

Article abstract of DOI:10.1016/S0040-4020(01)01114-0

The syntheses of 2-hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-alkanaminium 1(p) and 2-hydroxy-N,N-bis(2-hydroxyethyl)-N-methyl-1-alkanaminium 2(p) bromides are described for the homologous series, tetradecanaminium (p=14) through octadecaminium (p=18). The critical micelle concentrations (cmc's), measured by conductivity at 25 and 37°C, of the even-numbered members of the series show the expected linear dependence on chain length. All compounds display spermicidal and virucidal activity. Plots of minimum effective concentration (MEC) versus chain length in the Sander-Cramer spermicidal assay of 1 and 2 show a cut-off effect and a parabolic relationship, respectively. The best compounds - 1(17) and 1(18) - have MECs (～0.12mg mL^-1) slightly lower than that of the commercial spermicide nonoxynol-9 (N-9). Compounds 1(16) and 2(17) show the best anti-HIV activity in in-vitro, cell-free and cell-associated virus inactivation assays. Both are highly effective at 0.1mg mL^-1 in the cell-free assay and 0.5mg mL^-1 in the cell-associated assay. Compound 1(16) shows the best combination of dual activity against sperm and HIV. In summary, hydroxylated cationic surfactants, 1 and 2, show promise as inexpensive topical, contraceptive microbicides.

Full text of DOI:10.1016/S0040-4020(01)01114-0

TETRAHEDRON
Pergamon
Tetrahedron 58 (2002) 45±54
Spermicidal, anti-HIV, and micellar properties of di-
and trihydroxylated cationic surfactants
Yue-Ling Wong,a,² Christopher L. Curfman,a,³ Gustavo F. Doncel,b M. Patricia Hubieki,a,§
Travis C. Dudding,a,¶ Prashant S. Savlea,k and Richard D. Gandoura,p
aDepartment of Chemistry, Virginia Tech, Blacksburg, VA 24061-0212, USA
bDepartment of Obstetrics and Gynecology, CONRAD Program, Eastern Virginia Medical School, Norfolk, VA 23507, USA
Received 17 September 2001; accepted 12 November 2001
AbstractÐThe syntheses of 2-hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-alkanaminium 1(p) and 2-hydroxy-N,N-bis(2-hydroxyethyl)-
N-methyl-1-alkanaminium 2(p) bromides are described for the homologous series, tetradecanaminium (pˆ14) through octadecaminium
(pˆ18). The critical micelle concentrations (cmc's), measured by conductivity at 25 and 378C, of the even-numbered members of the series
show the expected linear dependence on chain length. All compounds display spermicidal and virucidal activity. Plots of minimum effective
concentration (MEC) versus chain length in the Sander±Cramer spermicidal assay of 1 and 2 show a cut-off effect and a parabolic
relationship, respectively. The best compoundsÐ1(17) and 1(18)Ðhave MECs (,0.12 mg mL21) slightly lower than that of the commer-
cial spermicide nonoxynol-9 (N-9). Compounds 1(16) and 2(17) show the best anti-HIV activity in in-vitro, cell-free and cell-associated virus
inactivation assays. Both are highly effective at 0.1 mg mL21 in the cell-free assay and 0.5 mg mL21 in the cell-associated assay. Compound
1(16) shows the best combination of dual activity against sperm and HIV. In summary, hydroxylated cationic surfactants, 1 and 2, show
promise as inexpensive topical, contraceptive microbicides. q 2002 Published by Elsevier Science Ltd.
Many women want to control their fertility and reduce their
risk of acquiring a sexually transmitted disease (STD).1±4
Developing user-controlled, topical vaginal microbicides
that provide protection against human immunode®ciency
virus type 1 (HIV-1) and other sexually transmitted patho-
gens is an urgent global priority.5
have met safety standards. The breakdown products of
N-9 and related compounds pose health and environmental
risks.18 Several research groups are searching for agents that
do not have these disadvantages.19±23
Quaternary ammonium salts have a long history as micro-
bicides24 and can act as spermicides; e.g. European and
Canadian spermicidal products contain benzalkonium
chloride, a popular microbicide. Although comparatively
less spermicidal than N-9,25 benzalkonium chloride has
potential as an anti-HIV, topical agent if the protection of
cynomolgus macaques against cervicovaginal trans-
mission26 of simian immunode®ciency virus (SIV)
mac251 translates to humans and HIV. Another common
microbicidal quaternary ammonium salt, N,N,N-trimethyl-
1-hexadecanaminium bromide, is also spermicidal,27 but has
not been developed further. Although simple quaternary
ammonium salts are quite microbicidal, they have some
disadvantages in that they have modest solubility in water
and can irritate sensitive tissue. Can one ®nd or synthesize
a quaternary ammonium salt that minimizes these short-
comings?
Nonoxynol-9 (N-9), currently the most widely used spermi-
cide,6 functions as a detergent by disrupting cell walls and
solubilizing membranes. It may increase the risk of HIV
transmission7 by activating interleukin-1-mediated NF-kB
activation, which leads to cytokine-induced recruitment of
HIV-1 host cells and increased HIV-1 replication.8 N-9
increases the risk of urinary tract infections9,10 vulvovaginal
candidiasis11 and genital ulcers.12 Furthermore, N-9, a
mixture of oligomers,13±17 may not meet future regulations
as the health-care industry increasingly uses pure
compounds or mixtures whose individual components
Keywords: spermicidal activity; anti-HIV activity; critical micelle concen-
tration; quaternary ammonium salts.
p
Corresponding author. Tel.: 11-540-231-3731; fax: 11-540-231-3255;
e-mail: gandour@vt.edu
Present Address: Department of Computer Science, Wake Forest U.,
Winston-Salem, NC, USA.
Present Address: Needle and Rosenberg, P.C., Atlanta, GA, USA.
Present Address: J-Star Research, Inc., South Plain®eld, NJ, USA.
Present Address: Department of Chemistry, Johns Hopkins U., Baltimore,
MD, USA.
Present Address: Rhodia ChiRex, Inc., Malvern, PA, USA.
²
Many reasons support our choice to investigate the water-
soluble quaternary ammonium salts, 1(p) and 2(p), as
spermicides and anti-HIV agents. They can be easily formu-
lated in hydrogels, common vehicles for the delivery of
spermicides. The 1(p) series, originally patented for anti-
static and fabric-softening ®nishes for textiles,28,29 is now
³
§
¶
k
0040±4020/02/$ - see front matter q 2002 Published by Elsevier Science Ltd.
PII: S0040-4020(01)01114-0

46
Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
Table 2. A and B values for equation that correlates cmc with chain length
(p) of 1 and 2 at 25 and 378C
Compounds and temperature
A
B
1 at 258C
1 at 378C
2 at 258C
2 at 378C
N,N,N-Trimethyl-1-alkanaminium
bromidesa at 608C
N,N,N-Trimethyl-1-alkanaminium
bromidesb at 258C
1.60
1.73
1.53
1.56
1.77
0.279
0.286
0.274
0.275
0.292
2.01
0.317
Scheme 1. (a) NBS, THF/H2O (5:1), rt, 18 h; (b) R1N(CH3)CH2CH2OH
(R1ˆCH3, CH2CH2OH), MeOH, heat, 3 d.
a
Refs. 36,34
Ref. 37
b
found in shampoo formulations,30 suggesting that it is a safe
for topical applications. A similar suggestion can be made
for the 2(p) series, which is patented for hair-conditioning
formulations.31 Furthermore, 1(p) and 2(p) are choline
derivatives and somewhat resemble an alkylated sphingo-
sine. Consequently, they might inhibit proteins involved in
cellular signaling and, thus, operate as selective micro-
bicides rather than indiscriminate detergents. Finally,
these compounds are made from inexpensive starting
materials, viz. hydroxyethylamines and 1-alkenes, which
are commodity chemicals.
1. Results
1.1. Chemistry
Reaction of an alkene 3a±e with N-bromosuccinimide32 in
THF/H2O gave a mixture of two bromohydrins 4 and 5 in
ratio of ca. 4:1. (Scheme 1.) The two isomers could be
separated by careful ¯ash column chromatography. Sepa-
ration of 4 and 5 was unnecessary, however, because
re¯uxing a mixture of either isomer with N,N-dimethyl-
ethanolamine (R1ˆCH3)28,33 and N-methyldiethanolamine
(R1ˆCH2CH2OH) in methanol gave quaternary ammonium
salts, 1 and 2, respectively. Pure quaternary ammonium salts
were isolated after careful work-up and puri®cation.
We used conductivity to measure the cmcs for selected
quaternary ammonium bromides, 1 and 2, respectively
(Table 1). As we anticipated, both 1 and 2 show an increase
in cmc with decrease in chain length.
One can determine the slope and intercept of the line
generated in a log cmc versus alkyl chain length at a
given set of conditions. The equation,34
In this paper, we report the experimental details for the
syntheses of both series and apply the historic method of
varying the chain-length (pˆ14±18) to probe structure-
activity in spermicidal and anti-HIV assays. Furthermore,
we have measured the critical micelle concentrations (cmcs)
of the even-numbered members of each series to test
whether these physicochemical properties affect the
structure±activity relationships.
log cmc ˆ A2BN;
where N represents the number of carbons in the long alkyl
chain attached to the quaternary nitrogen, enables com-
parisons among related surfactants. For 1 and 2, Nˆp. Inter-
cept A depends on the ionic strength of the system and on
Table 1. Critical micelle concentrations (cmcs) and alpha (ratio of slopes before and after the cmc) for 1 and 2 as a function of temperature and chain length as
measured by conductance in distilled water
Compound
103 cmc at 258C (mol dm23
)
103 cmc at 378C (mol dm23
)
Alpha at 258C
Alpha at 378C
1(14)
5.08^0.04
5.10^0.02
5.35^0.05
5.38^0.03
0.239^0.002
0.2399^0.0009
0.258^0.003
0.292^0.005
1(16)
1(18)
1.36^0.02
1.34^0.02
1.406^0.008
1.411^0.006
0.213^0.003
0.264^0.006
0.293^0.003
0.288^0.002
0.409^0.008
0.38^0.01
0.384^0.007
0.394^0.009
0.378^0.007
0.27^0.01
0.26^0.01
0.313^0.008
0.355^0.008
0.338^0.008
2(14)
2(16)
2(18)
5.06^0.04
5.07^0.02
5.28^0.02
5.26^0.02
0.253^0.003
0.260^0.002
0.296^0.003
0.299^0.002
1.26^0.01
1.262^0.009
1.351^0.009
1.37^0.05
0.235^0.003
0.252^0.002
0.285^0.002
0.31^0.01
0.409^0.009
0.401^0.007
0.422^0.006
0.413^0.006
0.200^0.004
0.254^0.003
0.308^0.005
0.332^0.003

Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
Table 3. Spermicidal activity of 1 in the Sander±Cramer assay
47
pˆ16, the values of cmcs rank in the following order:
N,N,N-trimethyl-1-hexadecanaminium bromide, 1.00 mM,
2(16) 1.26 mM,1(16), 1.35 mM. As replacement of methyl
by hydroxyethyl has no signi®cant effect on the cmc for
pˆ16,38 the cmc increases due to hydroxyl substitution on
the long alkyl chain.39 These equations also can accurately
predict the cmcs of an odd-numbered chain, e.g. the cmc of
1(15). The calculated value at 258CÐantilog(1.601
(0.279£15))ˆ2.60 mMÐmatches our measured valuesÐ
2.60^0.01 and 2.61^0.01 mM. At least for this limited
homologous series (pˆ14±18) of 1, the values of cmc's
maintain linearity in log-cmc-versus-alkyl-chain-length
plots.
Compound
Highest spermicidal
dilution (HSD) (1/X)
MEC (mg mL21
)
1(14)
1(15)
1(16)
1(17)
1(18)a
N-9b
34.0^4.4
117.3^14.8
186.7^33.3
192.0^18.5
192.0^18.5
72.0^7.7
0.781^0.163
0.202^0.023
0.140^0.019
0.117^0.011
0.117^0.011
0.156^0.016
Sperm samples (nˆ12) were mixed with two-fold serial dilutions of the
compounds in 0.9% NaCl (initial concentration of compound in distilled
water, 20 mg mL21) and observed under the microscope for 20 s. Those
dilutions that completely immobilized all screened spermatozoa were
further diluted in excess buffer and incubated for 1 h to verify lack of
motility recovery. The highest dilution that successfully passed both assess-
1.2. Biology
ments was considered the HSD and was used to calculate the MEC.
Requires warm water (408C) to remain in solution at 20 mg mL21
a
.
b
Initial concentration, 10 mg mL21
.
Table 3 presents the minimum effective concentrations
(MECs) for spermicidal activity in a modi®ed Sander±
Cramer assay40 for 1 and N-9. MECs represent a measure
of potency-the lower the MEC, the more potent the
compound. The Sander±Cramer assay determines the
minimum concentration required to stop all sperm motility
in 20 s. The MEC decreases; i.e. the potency increases, as
the chain length increases up to 1(17). Compounds 1(17)
and 1(18) have similar MECs, which are slightly lower than
that of N-9. Compound 1(18) has water-solubility problems;
hence, the value of its MEC may represent an upper limit.
Table 4. Spermicidal activity of 2 in the Sander±Cramer assay
Compound
HSD (1/X)
MEC (mg mL21
)
2(14)
2(15)
2(16)
2(17)a
2(18)a
N-9
30.7^3.5
45.3^4.6
62.7^7.2
92.0^11.0
56.0^10.1
72.0^7.7
0.378^0.043
0.248^0.022
0.202^0.040
0.156^0.042
0.235^0.028
0.156^0.016
Sperm samples (nˆ12) were mixed with two-fold serial dilutions of the
compounds in 0.9% NaCl (initial concentration of compound in distilled
water, 10 mg mL21) and observed under the microscope for 20 s. Those
dilutions that completely immobilized all screened spermatozoa were
further diluted in excess buffer and incubated for 1 h to verify lack of
motility recovery. The highest dilution that successfully passed both assess-
Table 4 presents MECs for spermicidal activity in a modi-
®ed Sander±Cramer assay for 2 and N-9. The MECs of the
series decrease as the chain length increases up to 2(17). The
MEC of 2(17) equals that of N-9. Compound 2(18) has
solubility problems; hence, the value of its MEC may
represent an upper limit.
ments was considered the HSD and was used to calculate the MEC.
Requires warm water (408C) to remain in solution at 10 mg mL21
a
.
Table 5 presents the data for anti-HIV activity of 1 in two
in-vitro assays: cell-free and cell-associated virus inacti-
vation. All chain lengths are effective, i.e. $3 log units of
infectious titer reduction at the two higher concentrations in
both assays. At the lower concentrations, 1(16) emerges as
the most potent against cell-free and cell-associated HIV.
Compound 1(16) has comparable activity to that of N-9 in
both assays.
the type of hydrophilic head group, and slope B depends on
the number of hydrophilic head groups and the number of
hydrophobic alkyl chains present in the amphiphile.35 Table
2 lists the A and B values for 1, 2, and N,N,N-trimethyl-1-
alkanaminium bromides.34,36,37
For this narrow range, temperature has little effect on the
cmc's of 1 and 2; the values of A and B are very similar at
25 and 378C. This limited comparison of three quaternary
ammonium bromides suggests that increasing hydroxylation
decreases A and B; both A and B decrease in the order
N,N,N-trimethyl-1-alkanaminium bromides.1.2. For
Table 6 presents the data for anti-HIV activity of 2 in the
cell-free and cell-associated HIV inactivation assays. All
chain lengths are effective, i.e. $3 log units of infectious
titer reduction at the two higher concentrations in both
Table 5. HIV cell-free and cell-associated inactivation assays for 1 and Nonoxynol-9 (N-9). Reduction of viral infectivity (in logs)
Cell-free inactivation assay concentration (mg mL21 Cell-associated inactivation assay concentration (mg mL21
)
)
Compound
1.0
0.5
0.1
2.0
1.6
3.0
2.3
2.0
3.3
0.05
1.5
1.3
2.3
1.3
1.6
1.5
1.0
0.5
0.1
0.05
1(14)
1(15)
1(16)
1(17)
1(18)
N-9
.3.7
.3.7
.3.7
.3.7
.3.7
.3.7
.3.7
.3.7
.3.7
.3.7
.3.7
.4.7
.4.5
.4.5
3.7
.4.5
.4.5
.4.5
.4.5
4.0
3.2
.4.5
.4.5
.4.5
,1.0
,1.0
3.0
,1.0
1.0
2.0
,1.0
,1.0
,1.0
,1.0
,1.0
1.4
Cell-free or cell-associated HIV-1 was incubated with multiple concentrations of the test agents for 2 min at 378C. Compound effect was then terminated by 10-
fold serial dilutions, and the treated virus was further incubated with infection-susceptible MT-2 cells for 6 d. Virus-induced cytopathic effects (formation of
cell syncytium) were recorded under microscopic observation. The values presented in the body of the table represent the reduction in the infectious titer of the
untreated virus (in logs) effected by each compound concentration.

48
Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
Table 6. HIV cell-free and cell-associated inactivation assays for 2 and nonoxynol-9 (N-9). Reduction of viral infectivity (in logs)
Cell-free inactivation assay concentration (mg mL21 Cell-associated inactivation assay concentration (mg mL21
)
)
Compound
1.0
0.5
0.1
0.05
1.0
0.5
0.1
0.05
2(14)
2(15)
2(16)
2(17)
2(18)
N-9
.4.5
.4.5
.4.5
.4.5
.4.5
.4.5
.5.5
.4.5
.4.5
.4.5
.4.5
.5.5
,1.0
,1.0
3.3
4.0
1.8
,1.0
,1.0
,1.0
1.0
,1.0
1.3
.3.8
.3.8
.3.8
.3.8
.3.8
.4.8
3.6
3.3
.3.8
.3.8
3.8
,0.3
0.3
,0.3
,0.3
,0.3
0.8
,0.3
,0.3
,0.3
,0.3
,0.3
,0.3
4.0
2.6
Cell-free or cell-associated HIV-1 was incubated with multiple concentrations of the test agents for 2 min at 378C. Compound effect was then terminated by 10-
fold serial dilutions, and the treated virus was further incubated with infection-susceptible MT-2 cells for 6 d. Virus-induced cytopathic effects (formation of
cell syncytium) were recorded under microscopic observation. The values presented in the body of the table represent the reduction in the infectious titer of the
untreated virus (in logs) effected by each compound concentration.
assays. At 0.5 mg mL21, 2(16) and 2(17) have a higher anti-
HIV activity than N-9 against the cell-associated virus.
Compounds 2(17) and 2(16) also represent the most potent
of the series against cell-free HIV, with comparable activity
to that of N-9.
In the curvilinear relationship between MEC and chain
length, the curve ¯attens for 1(17) and 1(18). This ¯attening
suggests chain length plays a role in the mechanism of
action. Three explanations come to mind. First, the result
might arise from two opposing effectsÐ(1) long chains
facilitate entry to the cell and lead to disruption of cellular
function and (2) long chains tend to form micelles at very
low concentrations. Such a decrease in cmc lowers the
concentration of monomer in solution. If the micelle does
not attack the cell, a higher total concentration of surfactant
is needed to have enough monomer to produce the bio-
logical result. A second explanationÐa chain-length selec-
tivity for incorporating the surfactant into the sperm
membraneÐis pure speculation. A third explanation is a
solubility problem, which is invisible to the naked eye,
with 1(18); there is less of it in solution and the MEC should
be lower.
2. Discussion
2.1. Spermicidal activity and micellar concentration
Fig. 1 presents the MECs for spermicidal activity and cmcs
for 1 plotted as a function of chain length. The cmcs are
expressed as mg mL21 to facilitate comparison with the
bioassays. The cmcs, which were measured in distilled
water, represent an upper limit of the cmc under the con-
ditions of the Sander±Cramer assay. That is, ,1% bovine
serum albumin (BSA), 0.9% saline, and Baker's (phosphate
and glucose) solution might decrease the cmc. BSA absorbs
cationic surfactants, e.g. benzalkonium;41 this absorption
should increase the total surfactant needed to form a
micelle. Salt signi®cantly decreases the cmc of hydroxy-
functionalized quaternary ammonium bromides.42 In the
presence of salt, the linear relationship between cmc and
chain length would likely shift to lower values.
Fig. 2 presents the MECs for spermicidal activity and cmcs
for 2 plotted as a function of chain length. As before, the
cmcs in water decrease linearly with increasing chain
length. When the cmcs are expressed in mg mL21
,
compounds 2 have higher cmcs than 1; the reverse occurs
when expressing cmcs in mol£dm23 (Tables 1 and 2). The
curvilinear relationship between MEC and chain length
Figure 1. Comparison of spermicidal activities (MECs) for 1 (B) and N-9
(Ð) with cmcs for 1 (V) in water.
Figure 2. Comparison of Spermicidal Activities (MECs) for 2 (B) and N-9
(Ð) with cmcs for 2 (V) in water.

Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
49
differs from that in Fig. 1. One must be careful, however, in
over-interpreting the MEC data because of potential solu-
bility problems. In Fig. 2, the MEC increases for 2(18)
compared to 2(17). The MEC for 2(17) represents a mini-
mum in the plot. This `parabola' in Fig. 2 contrasts with the
leveling off seen in Fig. 1 for the MECs of 1(17) and 1(18).
Compounds 1 and 2 might have different mechanisms for
biological activity, but an explanation based on micellar
effects or limited solubility cannot be excluded.
than, N-9. (2) The structural differences among the
compounds do not lead to large variations in anti-HIV
activity; these compounds inhibit HIV by similar mecha-
nisms. (3) Compound 1(16), which has excellent spermi-
cidal and anti-HIV properties, merits further investigation.
In summary, hydroxylated cationic surfactants, 1 and 2,
show promise as topical microbicides. A chain length of
16 carbons appears optimal for concomitant expression of
spermicidal and anti-HIV activity. Further structure-activity
studies on these compounds are warranted.
2.2. Anti-HIV properties
4. Experimental
4.1. General
For anti-HIV properties, (Tables 5 and 6) both 1 and 2 show
more activity against cell-free than the cell-associated
associated virus, an expected ®nding given the addition
`protection' offered by the cell carrier. Compounds 1(16),
2(16), and 2(17) are the best compounds in this study with
activities similar to or slightly better than that of N-9. The
data in Table 5 imply that 1(16) represents an optimum for
antiviral activity, as longer chain lengths are less active. The
data in Table 6 imply that 2(17) represents an optimum for
anti-HIV activity, as longer chain lengths are less active.
These results could arise from a biological effect related
to preferential membrane intercalation or a micellar effect.
Reagents were used as received. The solvents were HPLC
grade and used as received. Flash chromatography was
performed with silica gel, 60 X, 230±400 mesh. TLC was
performed with PE SIL G/UV 250 mm plates. Apparent
coupling constants, Japp, were the measured peak widths.
Atlantic MicroLab Inc. performed the elemental analyses.
4.2. General method for synthesis of bromohydrins 4a±e
and 5a±e
2.3. Comparisons with other work
To a cooled solution of a 1-alkene (10±15 mmol) in THF/
H2O (5:1) (ca. 40 mL g21 of alkene), was added slowly
N-bromosuccinimide (1.05 equiv.). The reaction mixture
was stirred at rt overnight. Hexanes (ca. same volume as
THF/H2O) was added, and the organic layer was washed
with water (3£25 mL), and then brine (50 mL). The organic
layer was dried over Na2SO4, ®ltered and concentrated
under reduced pressure to give a colorless oil. The crude
product was puri®ed by ¯ash column chromatography,
eluting with hexanes (2000 mL), then hexanes/ether (20:1;
400 mL; 10:1, 400 mL; 5:1, 400 mL, 2:1, until both bromo-
hydrins eluted), to give a mixture of the two isomeric
bromohydrins 4 and 5 (65±72% combined yield, about
4:1 ratio of 4:5). Small amounts of pure 4 and 5 were
isolated for analytical purposes.
As Shelton et al.43 found over a half century ago, the micro-
bicidal activity of a homologous series of N,N,N-trimethyl-
1-alkanaminium salts increases with increasing chain
length until a maximum occurs for the hexadecanaminium
homolog. More recent bioassays with a homologous series
of N,N,N-trimethyl-1-alkanaminium salts44,45 show similar
results. As Cella et al. suggested,46 this structure±activity
relationship may result from surface-active properties and
the tendency of these salts to form aggregates. Although the
mechanism of action is unknown, Hueck et al.'s47 early
suggestion that surface interactions of the quaternary am-
monium salt at the plasma membrane disrupt cellular
processes still holds promise. A recent study48 suggests
that quaternary ammonium salts do not reach the inner
monolayer of a cholesterol-rich bilayer membrane, e.g.,
the sperm plasma membrane. Incorporated surfactants
change the ¯uidity and porosity of a membrane; such
changes might have profound effects on cellular function.
1
4.2.1. 1-Bromo-2-tetradecanol (4a). H NMR (400 MHz,
CDCl3) d 3.77 (m, 1H), 3.54 (dd, 1H, Jappˆ10.2, 3.0 Hz),
3.38 (dd, 1H, Jappˆ10.2, 7.2 Hz), 2.11 (d, 1H, Jappˆ5.0 Hz),
1.56±1.25 (comp, 22H), 0.88 (t, 3H, Jappˆ6.9 Hz); (lit.32
1
The curvilinear relationships in Figs. 1 and 2 between
log MEC and chain length reveal the well-documented
60 MHz H NMR) 13C NMR (100 MHz, CDCl3) d 71.0,
40.7, 35.1, 31.9, 29.6, 29.58, 29.50, 29.45, 29.44, 29.3,
25.6, 22.6, 14.1; mass spectrum (CI) m/z 294, 291.1327
(M2H, C14H28BrO requires 291.1324).
cut-off effect49 and parabolic relationship,50 respectively.
51
Kopecky's recent review of these relationships might
Â
apply to our spermicidal data. Kopecky asserts, ª[cmc]¼
Â
1
4.2.2. 2-Bromo-1-tetradecanol (5a). H NMR (400 MHz,
represents the maximal available concentration of the
monomer¼[. This] may account for both the observed
decrease of the antimicrobial activity above [cmc] and¼the
cut-off effect of [very] hydrophobic [quaternary ammonium
salts] with low [cmcs]º.
CDCl3) d 4.14 (m, 1H), 3.81 (m, 1H), 3.74 (m, 1H), 2.09 (br
s, 1H), 1.82 (m, 2H), 1.57±1.20 (comp, 20H), 0.88 (t, 3H,
Jappˆ6.9 Hz); (lit.32 60 MHz 1H NMR) 13C NMR
(100 MHz, CDCl3) d 67.2, 60.1, 34.8, 31.9, 29.6, 29.58,
29.56, 29.48, 29.35, 29.29, 29.0, 27.4, 22.6, 14.1; mass
spectrum (CI) m/z 291.1329 (M2H, C14H28BrO requires
291.1324).
3. Conclusions
From these results, we draw three conclusions: (1) 1(17),
1(18), 1(16) and 2(17) display potent antispermatic activity
and are just as spermicidal as, or in some cases even more
4.2.3. 1-Bromo-2-pentadecanol (4b). 1H NMR (400 MHz,
CDCl3) d 3.77 (br m, 1H), 3.54 (dd, 1H, Jappˆ10.3, 3.2 Hz),
3.38 (dd, 1H, Jappˆ10.3, 7.1 Hz), 2.13 (d, 1H, Jappˆ3.5 Hz),

50
Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
1.56±1.25 (comp, 24H), 0.88 (t, 3H, Jappˆ6.8 Hz); 13C
NMR (100 MHz, CDCl3) d 71.0, 40.6, 35.1, 31.8, 29.63,
29.60, 29.58, 29.50, 29.45, 29.44, 29.3, 25.6, 22.6, 14.1;
mass spectrum (CI) m/z 307, 306, 305.1484 (M2H,
C15H30BrO requires 305.1480), 227, 213.
and N,N-dimethylethanolamine (2 equiv.) in CH3OH (about
1 mL/100 mg of the bromohydrin) was heated to re¯ux
overnight. The solvent was removed under reduced pressure
to give a crude product, which was puri®ed by recrystalli-
zation (conditions are speci®ed for individual compounds.).
4.2.4. 2-Bromo-1-pentadecanol (5b). 1H NMR (400 MHz,
CDCl3) d 4.15 (m, 1H), 3.82 (m, 1H), 3.75 (m, 1H), 2.01
(dd, 1H, Jappˆ7.9, 5.8 Hz), 1.84 (m, 2H), 1.55±1.26 (comp,
22H), 0.89 (t, 3H, Jappˆ6.9 Hz); 13C NMR (100 MHz,
CDCl3) d 67.2, 60.2, 34.8, 31.9, 29.63, 29.60, 29.56,
29.48, 29.35, 29.30, 29.0, 27.4, 22.6, 14.1; mass spectrum
(CI) m/z 307, 305.1469 (M2H, C15H30BrO requires
305.1480).
4.3.1. 2-Hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-
tetradecanaminium bromide, 1(14). Recrystallized in
EtOAc/CH2Cl2 (5:1) to give a white solid (85%): 1H
NMR (400 MHz, CDCl3) d 4.78 (t, 1H, Jappˆ5.3 Hz), 4.58
(d, 1H, Jappˆ7.0 Hz), 4.27 (br m, 1H), 4.10 (br m, 2H), 3.85
(m, 2H), 3.61 (dd, 1H, Jappˆ13.7, 10.1 Hz), 3.42 (comp,
7H), 1.57±1.23 (comp, 22H), 0.86 (t, 3H, Jappˆ6.9 Hz);
13C NMR (100 MHz, CDCl3) d 70.2, 66.7, 65.6, 55.9,
53.6, 53.3, 35.9, 31.9, 29.7, 29.64, 29.62, 29.57, 29.3,
25.2, 22.6, 14.1. Anal. calcd for C18H40NO2Br: C, 56.53;
H, 10.54; N 3.66. Found: C, 56.44; H, 10.64; N, 3.70.
1
4.2.5. 1-Bromo-2-hexadecanol (4c). H NMR (400 MHz,
CDCl3) d 3.77 (m, 1H), 3.54 (dd, 1H, Jappˆ10.3, 3.3 Hz),
3.37 (dd, 1H, Jappˆ10.3, 7.0 Hz), 2.19 (d, 1H, Jappˆ5.0 Hz),
1.57±1.19 (comp, 26H), 0.87 (t, 3H, Jappˆ6.9 Hz); 13C
NMR (100 MHz, CDCl3) d 71.0, 40.6, 35.1, 31.9, 29.65,
29.63, 29.62, 29.59, 29.51, 29.46, 29.44, 29.3, 25.6, 22.6,
14.1. (lit.52 90 MHz 1H NMR and 22.6 13C NMR).
4.3.2. 2-Hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-
pentadecanaminium bromide, 1(15). Recrystallized in
EtOAc/CH2Cl2 (5:1) to give a white solid (89%): 1H
NMR (400 MHz, CDCl3) d 4.80 (t, 1H, Jappˆ4.7 Hz), 4.59
(d, 1H, Jappˆ7.2 Hz), 4.27 (br m, 1H), 4.11 (br m, 2H), 3.85
(m, 2H), 3.62 (dd, 1H, Jappˆ13.4, 9.8 Hz), 3.42 (comp, 7H),
1.58±1.24 (comp, 24H), 0.87 (t, 3H, Jappˆ6.9 Hz); 13C
NMR (100 MHz, CDCl3) d 70.2, 66.7, 65.6, 56.0, 53.6,
53.3, 35.9, 31.9, 29.7, 29.64, 29.57, 29.3, 25.2, 22.6, 14.1.
Anal. calcd for C19H42NO2Br: C, 57.56; H, 10.68; N, 3.53.
Found: C, 57.25; H, 10.52; N, 3.67.
1
4.2.6. 2-Bromo-1-hexadecanol (5c). H NMR (400 MHz,
CDCl3) d 4.15 (m, 1H), 3.82 (m, 1H), 3.74 (m, 1H), 2.00
(dd, 1H, Jappˆ7.9, 5.8 Hz), 1.84 (m, 2H), 1.58±1.20 (comp,
24H), 0.88 (t, 3H, Jappˆ6.9 Hz); 13C NMR (100 MHz,
CDCl3) d 67.3, 60.2, 34.8, 31.9, 29.63, 29.60, 29.57,
29.49, 29.35, 29.32, 29.0, 27.4, 22.6, 14.1. (lit.53 R and S
1
enantiomers 400 MHz H NMR and 75 MHz 13C NMR).
4.3.3. 2-Hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-
hexadecanaminium bromide, 1(16). Recrystallized in
EtOAc/CH2Cl2 (2:1) to give a white solid (85%): 1H
NMR (400 MHz, CDCl3) d 4.79 (br t, 1H), 4.57 (d, 1H,
Jappˆ6.9 Hz), 4.28 (br m, 1H), 4.11 (br m, 2H), 3.86 (m,
2H), 3.62 (dd, 1H, Jappˆ13.5, 10.2 Hz), 3.43 (comp, 7H),
1.58±1.24 (comp, 26H), 0.89 (t, 3H, Jappˆ6.8 Hz); 13C
NMR (100 MHz, CDCl3) d 70.2, 66.7, 65.6, 56.0, 53.6,
53.3, 35.9, 31.9, 29.7, 29.64, 29.57, 29.3, 25.2, 22.6, 14.1.
Anal. calcd for C20H44NO2Br: C, 58.52; H, 10.80; N, 3.41.
Found: C 58.34, H 10.70, N 3.47.
4.2.7. 1-Bromo-2-heptadecanol (4d). 1H NMR (400 MHz,
CDCl3) d 3.78 (m, 1H), 3.55 (dd, 1H, Jappˆ10.3, 3.2 Hz),
3.38 (dd, 1H, Jappˆ10.3, 7.1 Hz), 2.09 (d, 1H, Jappˆ5.0 Hz),
1.57±1.25 (comp, 28H), 0.88 (t, 3H, Jappˆ6.9 Hz); 13C
NMR (100 MHz, CDCl3) d 71.0, 40.7, 35.1, 31.9, 29.64,
29.61, 29.58, 29.50, 29.43, 29.44, 29.3, 25.6, 22.6, 14.1.
4.2.8. 2-Bromo-1-heptadecanol (5d). 1H NMR (400 MHz,
CDCl3) d 4.15 (m, 1H), 3.82 (m, 1H), 3.74 (m, 1H), 2.00
(dd, 1H, Jappˆ7.9, 5.8 Hz), 1.84 (m, 2H), 1.56±1.25 (comp,
26H), 0.88 (t, 3H, Jappˆ6.9 Hz); 13C NMR (100 MHz,
CDCl3) d 67.3, 60.2, 34.8, 31.9, 29.63, 29.60, 29.56,
29.48, 29.35, 29.31, 29.0, 27.4, 22.6, 14.1.
4.3.4. 2-Hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-
heptadecanaminium bromide, 1(17). Recrystallized in
EtOAc/CH2Cl2/CH3OH (25:1:1) to give a white solid
1
4.2.9. 1-Bromo-2-octadecanol (4e). H NMR (400 MHz,
1
(83%): H NMR (400 MHz, CDCl3) d 4.79 (t, 1H, Japp
5.3 Hz), 4.56 (d, 1H, Jappˆ7.3 Hz), 4.29 (br m, 1H), 4.12 (br
ˆ
CDCl3) d 3.77 (m, 1H), 3.54 (dd, 1H, Jappˆ10.4, 3.2 Hz),
3.38 (dd, 1H, Jappˆ10.4, 7.1 Hz), 2.10 (d, 1H, Jappˆ5.0 Hz),
1.58±1.25 (comp, 30H), 0.88 (t, 3H, Jappˆ6.9 Hz); 13C
NMR (100 MHz, CDCl3) d 71.0, 40.6, 35.1, 31.9, 29.64,
29.60, 29.58, 29.5, 29.45, 29.43, 29.3, 25.6, 22.6, 14.1.
m, 2H), 3.87 (m, 2H), 3.63 (dd, 1H, Jappˆ13.4, 10.2 Hz),
3.43 (comp, 7H), 1.57±1.24 (comp, 28H), 0.87 (t, 3H, Japp
ˆ
6.9 Hz); 13C NMR (100 MHz, CDCl3) d 70.2, 66.7, 65.6,
56.0, 53.7, 53.3, 35.9, 31.9, 29.7, 29.63, 29.61, 29.55, 29.3,
25.2, 22.6, 14.1. Anal. calcd for C21H46NO2Br: C, 59.42; H,
10.92; N, 3.30. Found: C, 59.38; H, 10.98; N, 3.35.
1
4.2.10. 2-Bromo-1-octadecanol (5e). H NMR (400 MHz,
CDCl3) d 4.14 (m, 1H), 3.81 (m, 1H), 3.74 (m, 1H), 2.07
(dd, 1H, Jappˆ7.8, 5.8 Hz), 1.84 (m, 2H), 1.57±1.22 (comp,
28H), 0.88 (t, 3H, Jappˆ6.8 Hz); 13C NMR (100 MHz,
CDCl3) d 67.2, 60.2, 34.8, 31.9, 29.65, 29.61, 29.57, 29.5,
29.4, 29.3, 29.0, 27.4, 22.6, 14.1.
4.3.5. 2-Hydroxy-N-(2-hydroxyethyl)-N,N-dimethyl-1-
octadecanaminium bromide, 1(18). Recrystallized in
EtOAc to give a white solid (88%): H NMR (400 MHz,
1
CDCl3) d 4.80 (t, 1H, Jappˆ5.4 Hz), 4.57 (d, 1H, Japp
ˆ
7.0 Hz), 4.30 (br m, 1H), 4.12 (br m, 2H), 3.87 (m, 2H),
3.63 (dd, 1H, Jappˆ13.4, 9.9 Hz), 3.43 (comp, 7H), 1.59±
1.19 (comp, 30H), 0.87 (t, 3H, Jappˆ6.9 Hz); 13C NMR
(100 MHz, CDCl3) d 70.2, 66.7, 65.6, 56.0, 53.6, 53.3,
35.9, 31.9, 29.7, 29.6, 29.5, 29.3, 25.2, 22.6, 14.1). Anal.
4.3. General methods for synthesis of quaternary
ammonium bromides, 1(14±18)
A solution of bromohydrin (4 or 5 or a mixture of 4 and 5)

Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
51
calcd for C22H48NO2Br: C, 60.26; H, 11.03; N, 3.19. Found:
C, 60.12; H, 10.87; N, 3.26.
vial; rinsed with CHCl3 (about 0.2 mL). EtOAc (10 mL)
was added and the solution turned milky. After standing
for 1 h at rt, ®ne crystals formed. The white crystals were
collected by vacuum ®ltration and dried under high vacuum
for 3 d at rt. 1H NMR (400 MHz, CDCl3) d 4.72 (two over-
lapping triplets, 2H), 4.48 (d, 1H, Jappˆ7.2 Hz),4.27 (m,
1H), 4.12 (m, 4H), 3.94±3.86 (comp, 4H), 3.67 (dd, 1H,
Jappˆ13.3, 10.1 Hz), 3.46 (d, 1H, Jappˆ13.3 Hz), 3.41 (s,
3H3), 1.58±1.20 (comp, 26H), 0.88 (t, 3H, Jappˆ6.9 Hz);
13C NMR (100 MHz, CDCl3) d 68.3, 65.7, 65.2, 64.9,
55.9, 51.2, 36.1, 31.9, 29.82, 29.79, 29.76, 29.69, 29.4,
25.3, 22.7, 14.1. Anal. calcd for C21H46NO3Br: C, 57.26;
H, 10.53; N, 3.18. Found: C, 57.29; H, 10.53; N, 3.17.
4.4. General method for the synthesis of quaternary
ammonium bromides, 2(14±18)
A solution of bromohydrin (a mixture of 4 and 5) and
N-methyldiethanolamine (1.8 equiv.) in CH3OH (about
1 mL/100 mg of the bromohydrin) was heated to re¯ux for
3 d. The reaction mixture was concentrated to give pale
yellow oil. Repeated trituration of the oil with EtOAc
(25 mL g21 of oil/treatment) eventually produced a white
solid. (Care was taken to break up lumps during trituration!)
The crude solid was puri®ed by recrystallization (conditions
are speci®ed for individual compounds.).
4.4.4. 2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-methyl-1-
heptadecanaminium, bromide, 2(17). The solid
(,200 mg) was dissolved in warm CHCl3 (,0.5 mL).
The solution turned milky after addition of CH3CN
(,0.5 mL) and then EtOAc (7 mL). Crystals were formed
after 4 h; the solution was allowed to stand at rt overnight.
The crystals were collected by vacuum ®ltration, washed
with EtOAc (20 mL), air-dried on the Buchner funnel for
4.4.1. 2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-methyl-1-
tetradecanaminium, bromide, 2(14). The solid
(,200 mg) was dissolved in warm CHCl3 (,0.5 mL). The
solution turned milky after addition of CH3CN (,0.5 mL)
and then EtOAc (7 mL). Crystals were formed after 4 h; the
solution was allowed to stand at rt overnight. The crystals
were collected by vacuum ®ltration, washed with EtOAc
(20 mL), air-dried on the Buchner funnel for 10 min, and
then dried under high vacuum for 3 d at rt. 1H NMR
(400 MHz, CDCl3) d 4.64 (two overlapping triplets, 2H),
4.43 (d, 1H, Jappˆ7.2 Hz), 4.28 (m, 1H), 4.12 (m, 4H),
3.96±3.86 (comp, 4H), 3.68 (dd, 1H, Jappˆ13.7, 9.9 Hz),
3.47 (d, 1H, Jappˆ13.7 Hz), 3.41 (s, 3H), 1.54±1.25
(comp, 22H), 0.88 (t, 3H, Jappˆ6.9 Hz); 13C NMR
(100 MHz, CDCl3) d 68.4, 65.7, 65.3, 65.0, 55.9, 51.2,
36.1, 31.9, 29.8, 29.7, 29.6, 29.4, 25.3, 22.7, 14.1. Anal.
calcd for C19H42NO3Br: C, 55.33; H, 10.26; N, 3.40.
Found: C, 55.20; H, 10.32; N, 3.35.
1
10 min, and then dried under high vacuum for 3 d at rt. H
NMR (400 MHz, CDCl3) d 4.76 (two overlapping triplets,
2H), 4.52 (d, 1H, Jappˆ6.6 Hz),4.26 (m, 1H), 4.11 (m, 4H),
3.93±3.84 (comp, 4H), 3.66 (dd, 1H, Jappˆ12.6, 10.5 Hz),
3.46 (d, 1H, Jappˆ12.6 Hz), 3.39 (s, 3H3), 1.58±1.20 (comp,
28H), 0.88 (t, 3H, Jappˆ6.9 Hz); 13C NMR (100 MHz,
CDCl3) d 68.3, 65.7, 65.3, 65.0, 55.9, 55.8, 51.2, 36.1,
31.9, 29.81, 29.77, 29.75, 29.67, 29.4, 25.3, 22.7, 14.1.
Anal. calcd for C22H48NO3Br: C, 58.13; H, 10.64; N, 3.08.
Found: C, 57.99; H, 10.59; N, 3.14.
4.4.5. 2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-methyl-1-
octadecanaminium, bromide, 2(18). The solid was
dissolved in boiling CH2Cl2/CHCl3/EtOAc (20/10/10 mL),
®ltered through a ®lter paper. The ®ltrate was cooled at rt for
30 min; a white solid was collected by vacuum ®ltration,
4.4.2. 2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-methyl-1-
pentadecanaminium, bromide, 2(15). The white solid
(,200 mg) was dissolved in warm CHCl3 (,0.3 mL) and
CH3CN (about 0.5 mL) and EtOAc (10 mL) were added.
The resulting milky solution remained at rt for 12 h. The
white solid was collected by vacuum ®ltration, and washed
with EtOAc/CH3CN (5:1, 10±15 mL). The white solid was
air-dried on the Buchner funnel for 10 min, but the white
solid was still too sticky to scrap out of the ®lter paper. The
whole ®lter paper with the solid was dried in vacuo over-
night in a dessicator. The solid was scrapped out carefully,
and dried under high vacuum for 3 d at rt. 1H NMR
(400 MHz, CDCl3) d 4.67 (two overlapping triplets, 2H),
4.46 (d, 1H, Jappˆ6.4 Hz),4.26 (m, 1H), 4.11 (m, 4H), 3.91±
3.86 (comp, 4H), 3.67 (dd, 1H, Jappˆ13.0, 9.3 Hz), 3.47 (d,
1H, Jappˆ13.0 Hz), 3.41 (s, 3H3), 1.53±1.25 (comp, 24H),
0.88 (t, 3H, Jappˆ6.4 Hz); 13C NMR (100 MHz, CDCl3) d
68.4, 65.7, 65.3, 65.0, 55.9, 51.2, 36.1, 31.9, 29.8, 29.74,
29.70, 29.66, 29.4, 25.3, 22.7, 14.1. Anal. calcd for
C20H44NO3Br: C, 56.33; H, 10.40; N, 3.28. Found: C,
56.19; H, 10.45; N, 3.28.
1
and washed with CH2Cl2 (20 mL). H NMR (400 MHz,
CDCl3) d 4.72 (two overlapping triplets, 2H), 4.48 (d, 1H,
Jappˆ7.0 Hz), 4.27 (m, 1H), 4.11 (m, 4H), 3.97±3.85 (comp,
4H), 3.67 (dd, 1H, Jappˆ13.2, 10.1 Hz), 3.45 (d, 1H, Japp
ˆ
13.2 Hz), 3.40 (s, 3H3), 1.58±1.20 (comp, 30H), 0.88 (t, 3H,
Jappˆ6.5 Hz); 13C NMR (100 MHz, CDCl3) d 68.3, 65.7,
65.3, 65.0, 55.9, 55.8, 51.2, 36.1, 31.9, 29.80, 29.76,
29.67, 29.4, 25.3, 22.7, 14.1. Anal. calcd for
C23H50NO3Br: C, 58.96; H, 10.76; N, 2.99. Found: C,
58.66; H, 10.55; N, 3.06.
4.5. Determination of critical micelle concentrations by
conductivity
Conductivity measurements were made with a conductance
meter connected to a conductance cell, containing plati-
num±iridium electrodes coated with platinum black and
gold soldered to platinum lead wires, a dip-type cell made
from ABS plastic with a cgs cell constant of 1.0 cm21. The
meter gave readings in siemens cm21 (S cm21). A tempera-
ture probe connected directly to the meter measured the
temperature of the sample solution before and after each
set of data was collected. To prevent sample polarization
inside the cell, the conductance meter operated with a
50±60 Hz AC circuit. The factory-set calibration of the
4.4.3. 2-Hydroxy-N,N-bis(2-hydroxyethyl)-N-methyl-1-
hexadecanaminium, bromide, 2(16). The white solid
(,400 mg) was dissolved in warm CHCl3 (1 mL) and
CH3CN (1 mL) was added. The solution was warmed to
dissolve the entire solid and was ®ltered through a small
amount of ¯ash silica gel in a 600 pipet into a wide-mouth

52
Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
Figure 3. Typical Plot of conductivity versus concentration of 1(16). Lines represent linear regressions through consecutive data. The equation yˆmbx1bb are
data below the cmc, yˆmax1ba are data above the cmc.
conductance meter and the integrity of the conductivity cell
were checked periodically by with two NIST traceable
conductivity standards (13.61 and 102.3 mS cm21).
the determined apparent density gave the volume of added
stock solution. (2) Adding the total volume of the added
sample to the original sample volume gave the volume of
the sample solution. (3) Multiplying the total delivered
volume by the stock-solution concentration, and dividing
by the sample volume gave the concentration of the sample.
In a plot of conductivity versus sample concentration, the
cmc was taken as the point where the slope of the line
drastically changes. Fig. 3 is a plot of typical data.
Conductivity measurements were made on a 25-mL,
sample-volume scale. Prior to use, all glassware was washed
and rinsed with distilled H2O and acetone, then dried in an
oven at 1008C for 6 h. The sample beaker was immersed in a
water bath whose temperature was controlled with an
immersion circulator. The sample was continuously stirred
with a magnetic stir bar. All weights were measured with an
analytical balance (readability 0.01 mg) that was internally
calibrated before use.
The graph of concentration vs. conductivity can be thought
of as two separate straight lines, illustrated above. The
equations yˆmax1ba and yˆmbx1bb, where the subscripts
`b' and `a', respectively, stand for `before' and `after' the
break in slope, describe these lines. At the cmc, taken as the
point where the two lines intersect, (yb,xb)ˆ(ya,xa). There-
fore, the equations for the two lines can be set equal to one
another and the x-coordinate of the intersection point (the
cmc) solved, mbx 1 bbˆmax 1 ba or xˆ(ba2bb)/(mb2ma).
Weighing out the compound in volumetric ¯asks and
diluting to the mark with distilled H2O provided stock solu-
tions, whose concentrations ranged from 0.756 to 0.0708 M;
however, the majority of the stock solution concentrations
were between 0.1 and 0.2 M. The weight of the solution was
recorded; the apparent density was calculated for the current
rt, typically between 19 and 248C. (There are errors asso-
ciated with these density calculations because the glassware
was calibrated for 208C. However, because of the 0.11%
difference between the density of water at 19 and 248C,
we neglected the errors in the volume measurements asso-
ciated with variations in rt.) A volumetric pipette was used
to add distilled H2O to the sample beaker, which was then
placed into a water bath. The conductance cell was placed in
the H2O and the system equilibrated for ,5 min. The purity
of the H2O was evaluated by the conductivity, giving
Linear regression analyses were performed to determine ba,
bb, mb, and ma. As the cmc is actually a narrow concentration
range and not a single speci®c point, the lines curved
slightly before and after the cmc. To increase the robustness
of the results, several linear-regression analyses were
performed. The initial regressions used the ®ve data points
furthest from the in¯ection, and then, adding one point at a
time, approaching the in¯ection. To determine which
regression data were retained, the standard error of each
regression analysis was calculated. If the standard error
for a regression was more than three times the initial ®ve-
point regression's standard error, that set of data was
rejected. In this way variations due to the gradual curving
of the lines near the cmc were avoided. Then every possible
combination of the regression data was used to calculate all
possible cmc's. For example, if lines calculated with 5, 6, 7,
and 8 data points ®tted the criteria both before and after
the cmc, then 16 (4£4) calculations of the cmc resulted.
The means of these calculations, along with the standard
deviation from the mean, were reported in Table 1.
readings no higher than 3.00 mS cm21
.
The amount of stock solution delivered to the beaker was
measured by weighing a 250-mL syringe before and after
sample delivery. The solution equilibrated for ,10 min
before recording the conductivity reading and adding the
next aliquot of stock solution. The procedure was repeated
until 20 data points (typically 10 below and 10 above the
cmc) were collected. Data points up to but not exceeding
®ve times the cmc were collected.
The weight of the stock solution added to the sample was
calculated by taking the difference of the syringe weights
before and after injection. These weights were entered into a
Microsoftw Excel spread sheet where the results were calcu-
lated by the following method: (1) dividing the weight by
4.6. Semen samples
Semen samples were collected by healthy volunteers by
masturbation and allowed to liquefy for 30 min at rt. If com-
pletely lique®ed, they were evaluated for sperm concentration

Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
53
and motion parameters with a computer-assisted semen
analyzer (Hamilton Thorne Research, Beverly, MA) Only
specimens with .60£106 motile sperm/mL and .50%
motility were used.
the use of the conductivity meter and Professor Thomas C.
Ward for the thermostat bath.
References
4.7. In vitro spermicidal activity
1. Hardy, E.; de Padua, K. S.; Osis, M. J.; Jimenez, A. L.;
Zaneveld, L. J. Contraception 1998, 58, 251±255.
2. Hardy, E.; Jimenez, A. L.; de Padua, K. S.; Zaneveld, L. J.
Contraception 1998, 58, 245±249.
Spermicidal activities of the compounds were evaluated at
once using a modi®ed version to the protocol originally
described by Sander and Cramer.40 Semen samples were
adjusted with
a phosphate-buffered glucose solution
3. Hardy, E.; de Padua, K. S.; Jimenez, A. L.; Zaneveld, L. J.
Contraception 1998, 58, 239±244.
(Baker's buffer) supplemented with 0.5±1.0% BSA in
order to contain 60£106 motile sperm/mL; two-fold serial
dilutions of the compounds were prepared in 0.9% NaCl
(saline). Fifty microliters of semen were pipetted into
250 mL of compound/saline dilutions and gently vortexed
for 3 s. A drop of this mixture was placed on a pre-warmed
glass slide, coverslipped, and analyzed under a dark-®eld
microscope with a 10£ objective during 30 s. If any motile
spermatozoon was seen, the dilution was considered as
negative. Positive dilutions, i.e. all observed sperm immo-
tile, were further incubated at 378C for 1 h with two volumes
of Baker's buffer, and then re-examined for sperm motility.
If no motile sperm were seen, the positive score was main-
tained. The MEC of a compound was calculated with the
highest sperm-immobilizing dilutions and the initial
concentration of the compound. Several different donor
semen samples were assayed to ensure consistency of
results. Nonoxynol-9 (N-9) was used as positive control in
all assays.
4. Hardy, E.; de Padua, K. S.; Jimenez, A. L.; Zaneveld, L. J.
Contraception 1998, 58, 233±238.
5. McCormack, S.; Hayes, R.; Lacey, C. J. N.; Johnson, A. M.
BMJ 2001, 322, 410±413.
6. Coggins, C.; Elias, C. Int. J. Gynaecol. Obstet. 2000, 68, 267±
268.
7. Stephenson, J. JAMA 2000, 284, 949.
8. Fichorova, R. N.; Tucker, L. D.; Anderson, D. J. J. Infect. Dis.
2001, 184, 418±428.
9. Fihn, S. D.; Boyko, E. J.; Chen, C. L.; Normand, E. H.;
Yarbro, P.; Scholes, D. Arch. Intern. Med. 1998, 158, 281±
287.
10. Fihn, S. D.; Boyko, E. J.; Normand, E. H.; Chen, C. L.;
Grafton, J. R.; Hunt, M.; Yarbro, P.; Scholes, D.; Stergachis,
A. Am. J. Epidemiol. 1996, 144, 512±520.
11. Geiger, A. M.; Foxman, B. Epidemiology 1996, 7, 182±187.
12. Feldblum, P. J. Genitourin. Med. 1996, 72, 451±452.
13. Black, D. B.; Dawson, B. A.; Ethier, J. C.; Neville, G. A.
J. Pharm. Biomed. Anal. 1990, 8, 527±533.
4.8. Statistical analysis
14. Zavos, P. M.; Correa, J. R.; Nosek, D.; Mohammadi, F.;
Digenis, G. A. Fertil. Steril. 1996, 66, 729±733.
15. Zavos, P. M.; Correa, J. R.; Nosek, D.; Mohammadi, F.;
Digenis, G. A. Contraception 1996, 54, 39±41.
16. Walter, B. A.; Digenis, G. A. Pharm. Res. 1991, 8, 409±411.
17. Walter, B. A.; Hawi, A. A.; Zavos, P. M.; Digenis, G. A.
Pharm. Res. 1991, 8, 403±408.
Comparisons between pairs of spermicidal results were
done with the non-parametric Mann±Whitney test.
4.9. In vitro anti-HIV activity
Following the method of Resnick et al.54 cell-free (RF
strain) and cell-associated (RF-infected H9 cells) HIV-1
stocks, displaying titers around 62log 10 TCID50, were
incubated with four dilutions of the test agents (starting at
0.1%) for 2 min. Compound effect was terminated by
10-fold serial dilutions, which were then incubated with
plated MT-2 cells for 4±6 d (in quadruplicates). Endpoint:
syncytium formation and compound cytotoxicity
(agents1MT-2 cells without virus). Results were calculated
by the Reed±Muench method.55 N-9 was used as positive
control.
18. Scott, M. J.; Jones, M. N. Biochim. Biophys. Acta-Biomembr.
2000, 1508, 235±251.
19. Savle, P. S.; Doncel, G. F.; Bryant, S. D.; Hubieki, M. P.;
Robinette, R. G.; Gandour, R. D. Bioorg. Med. Chem. Lett.
1999, 9, 2545±2548.
20. Gagne, N.; Cormier, H.; Omar, R. F.; Desormeaux, A.;
Gourde, P.; Tremblay, M. J.; Juhasz, J.; Beauchamp, D.;
Rioux, J. E.; Bergeron, M. G. Sex. Transm. Dis. 1999, 26,
177±183.
21. Belec, L.; Tevi-Benissan, C.; Bianchi, A.; Cotigny, S.;
Beumont-Mauviel, M.; Si-Mohamed, A.; Malkin, J. E. J. Anti-
microb. Chemother. 2000, 46, 685±693.
22. D'Cruz, O. J.; Waurzyniak, B.; Yiv, S. H.; Uckun, F. M.
Contraception 2000, 61, 69±76.
Acknowledgements
23. Krebs, F. C.; Miller, S. R.; Catalone, B. J.; Welsh, P. A.;
Malamud, D.; Howett, M. K.; Wigdahl, B. Antimicrob. Agents
Chemother. 2000, 44, 1954±1960.
This work was supported by contracts from the Contracep-
tive Research and Development (CONRAD) program,
Eastern Virginia Medical School, under a Cooperative
Agreement with the US Agency for International Develop-
ment (USAID). The views expressed by the authors do not
necessarily re¯ect those of CONRAD or USAID. We thank
Personal Products Company, a division of Ortho-McNeil
Pharmaceutical, Inc. (Raritan, NJ) for generously providing
samples of N-9. We thank Professor Richey M. Davis, for
24. Merianos, J. J. Quaternary Ammonium Antimicrobial
Compounds. In Disinfection, Sterilization, and Preservation,
Block, S. S., Ed.; 4th ed, Lea and Febiger: Malvern, 1991;
pp 225±255.
25. Furuse, K.; Ishizeki, C.; Iwahara, S. J. Pharmacobiodyn. 1983,
6, 359±372.
26. Tevi-Benissan, C.; Makuva, M.; Morelli, A.; Georges-Courbot,

54
Y.-L. Wong et al. / Tetrahedron 58 (2002) 45±54
M. C.; Matta, M.; Georges, A.; Belec, L. J. AIDS 2000, 24,
147±153.
40. Sander, F. V.; Cramer, S. D. Hum. Fertil. 1941, 6, 134±137.
41. Jono, K.; Takayama, T.; Kuno, M.; Higashide, E. Chem.
Pharm. Bull. 1986, 34, 4215±4224.
27. Chow, P. Y.; Holland, M. K.; Suter, D. A.; White, I. G. Int. J.
Fertil. 1980, 25, 281±286.
42. Wilk, K. A.; Burczyk, B.; Kubica, H. Colloids Surf. 1990, 50,
363±371.
28. Rutzen, H. Process for the Preparation of Quaternary Ammo-
nium Compounds, (Henkel, K.-G.a.A., Fed. Rep. Ger.). US
Patent 4,480,126, October 30, 1984.
43. Shelton, R. S.; Van Campen, M. G.; Tilford, C. H.; Lang,
H. C.; Nisonger, L.; Bandelin, F. J.; Rubenkoeng, H. L.
J. Am. Chem. Soc. 1946, 68, 753±755.
29. Rutzen, H.; Nikolaus, P.; Bischoff, M.; Lehmann, R. Process
for Manufacture of Quaternary Ammonium Compounds;
(Henkel, K.-G.a.A., Fed. Rep. Ger., Degussa, A.G.). US Patent
4,492,802, January 8, 1985.
30. Henkel KGaA, Dehyquartw E-CA; http://surfactants.net/
formulary/produkte_rezepturen/dehyquart_e_ca.htm
(accessed August, 2001).
44. Majtan, V.; Majtanova, L.; Mlynarcik, D. Arzneim.-Forsch.
1995, 45, 198±199.
45. Wright, N. E.; Gilbert, P. J. Pharm. Pharmacol. 1987, 39,
685±690.
46. Cella, J. A. E. D. N.; Noel, D. R.; Harriman, L. A.; Harwood,
H. J. J. Am. Chem. Soc. 1952, 74, 2061±2061.
47. Hueck, H. J.; Adema, D. M.; Wiegmann, J. R. Appl. Micro-
biol. 1966, 14, 308±319.
31. Hensen, H.; Rutzen, H.; Busch, P.; Stuhrmann, D.; Thiele, K.
Quaternary Ammonium Compound Hair Conditioners,
(Henkel K.-G.a.A., Fed. Rep. Ger.). US Patent 4,744,977,
May 17, 1988; 15 pp.
48. Watanabe, S.; Regen, S. L. J. Am. Chem. Soc. 1994, 116,
5762±5765.
49. Devinsky, F.; Kopecka-Leitmanova, A.; Sersen, F.; Balgavy,
P. J. Pharm. Pharmacol. 1990, 42, 790±794.
50. Hansch, C.; Clayton, J. M. J. Pharm. Sci. 1973, 62, 1±21.
Â
51. Kopecky, F. Pharmazie 1996, 51, 135±144.
52. Kotsuki, H.; Shimanouchi, T.; Ohshima, R.; Fujiwara, S.
Tetrahedron 1998, 54, 2709±2722.
32. Hanzlik, R. P.; Edelman, M.; Michaely, W. J.; Scott, G. J. Am.
Chem. Soc. 1976, 98, 1952±1955.
33. Broxton, T. J.; Chung, R. P. T. J. Org. Chem. 1990, 55, 3886±
3890.
34. Klevens, H. B. J. Am. Oil Chem. Soc. 1953, 30, 74±80.
35. Shinoda, K. The Formation of Micelles. Colloidal
Surfactants: Some Physicochemical Properties; Shinoda, K.,
Nakagawa, T., Tamamushi, B. I., Isemura, T., Eds.;
Academic: New York, 1963; Vol. 12, pp 1±96.
36. Scott, A. B.; Tartar, H. V. J. Am. Chem. Soc. 1943, 65, 692±
698.
53. Guo, J.; Knapp, D. W.; Boegeman, S. Tetrahedron:
Asymmetry 2000, 11, 4105±4111.
54. Resnick, L.; Busso, M. E.; Duncan, R. C. Anti-HIV Screening
Technology. In Alexander, N. J., Gabelnick, H. L., Spieler, J.
M. (Eds.), Heterosexual Transmission of AIDS, Proceedings
of the Second Contraceptive Research and Development
(CONRAD) Program International Workshop, held in
Norfolk, Virginia, February 1±3, 1989; Wiley-Liss: New
York, 1990; pp 311±325.
37. Zana, R. J. Colloid Interface Sci. 1980, 78, 330±337.
38. Ralston, A. W.; Eggenberger, D. N.; Harwood, H. J.; Du
Brow, P. L. J. Am. Chem. Soc. 1947, 69, 2095±2097.
39. Burczyk, B.; Wilk, K. A. Prog. Colloid Polym. Sci. 1990, 82,
249±252.
55. Reed, L. J.; Muench, H. Am. J. Hyg. 1938, 27, 493±497.