Content uploaded by Nathalie Bourgougnon
Author content
All content in this area was uploaded by Nathalie Bourgougnon on Jan 10, 2014
Content may be subject to copyright.
443
ISSN 0326-2383
KEY WORDS: Antiviral, Carrageenan, Herpes Simplex viruses, Poliovirus, Polysaccharide, Red algae.
* Author to whom correspondence should be addressed. E-mail: jarbas.montanha@farmacia.ufrgs.br
Latin American Journal of Pharmacy
(formerly Acta Farmacéutica Bonaerense)
Lat. Am. J. Pharm. 28 (3): 443-8 (2009)
Original Article
Received: August 26, 2008
Accepted: March 14, 2009
Antiviral Activity of Carrageenans from Marine Red Algae
Jarbas A. MONTANHA 1*, Nathalie BOURGOUGNON 2, Joel BOUSTIE 3& Maryvonne AMOROS 3
1Faculdade de Farmácia da Universidade Federal do Rio Grande do Sul,
Av. Ipiranga 2752 (UFRGS), Porto Alegre, RS CEP 90610-000, Brazil;
2Laboratoire de Biotechnologie et Chimie Marines Université de Bretagne Sud Centre de
Recherche et d’Enseignement Yves Coppens Campus de Tohannic, BP 573- 56017 Vannes, France;
3Laboratoire dês Substances Licheniques et Photoprotection,
Equipe Pharmacognosie et Mycologie, Faculté de Pharmacie, Université de Rennes I,
2 Avenue du Prof. L. Bernard,35043, Rennes Cedex, France.
SUMMARY. Three carrageenan representatives of each structural type: λ- and ι–family (Gigartina acicu-
laris), ι-family (Euchema denticulatum) and κ–family (Kappaphycus cottonii) have been tested for their in
vitro antiviral activity. The carrageenans proved to be potent inhibitors of herpes human virus type 1
(HHV-1) and Poliovirus. The best results were obtained with carrageenans from Gigartina acicularis and
Euchema denticulatum, which are more sulfated than those from Kappaphycus cottonii. The selective index
values (CC50/ID50) ranged from more than 22 to more than 545 for HHV-1 and more than 6.6 to more
than 32 for Poliovirus. No citotoxic effects were observed. At 0.75 mg/ml, none of the carrageenans tested
showed a virucidal activity against HHV-1 or Poliovirus. Carrageenans from Euchema denticulatum (ι-
family) and Gigartina acicularis (λ- and ι–family) exerted their antiviral effect via, in part, by a lower in-
hibition of the virus attachment and by the interference in a subsequent stage of the virus replicative cycle.
The κ-carrageenan from Kappaphycus cottonii exerted its antiviral effect mainly by a lower inhibition of
the virus attachment. In cultures treated with carrageenans from Euchema denticulatum (ι-family) and Gi-
gartina acicularis (λ- and ι–family), the HHV-1 viral DNA synthesis had a reduction of threefold and
twofold with 0.75 mg/ml, respectively.
INTRODUCTION
Carrageenans are sulfated polydigalactosides
(20-50% -OSO3Na) with molecular weight be-
tween 105and 106that can be extracted from
several red seaweeds. They comprise a broad
range of structures and are divided into families:
the κ-family is defined by the presence of a C-4-
sulfate group on the β-D-galactose unit; the ι-
family is characterized by a C-4-sulfate group on
the β-D-galactose unit and a C-2-sulfate group
on the 3-6-anhydro-α-D-galactose; the λ-family
is characterized by a sulfate group on the C-2 of
β-D-galactose unit and 2,6-dissulfate on α-D-
galactose unit 1. Polysaccharides is a complex
group of peculiar molecules exhibiting a wide
range of biological activities such as antiinflam-
matory, anticoagulant, antithrombotic, antitu-
moral, antimetastatic, antifertilizing and antiviral
2. Unlike antimicrobial drugs against bacteria
and fungi, only a few effective antiviral drugs
are available. One of the most important rea-
sons for the lack of success in developing an-
tiviral drugs is due to the nature of the infec-
tious viral agents, which totally depend upon
the cell they infect for multiplication and sur-
vival, so, compounds that may cause the death
of viruses also are very likely to injure the host
cell that harbour them 3.
Efforts have been made to evaluate the an-
tiviral activity of natural products, including
those from algaes, in order to characterize new
compounds which could inhibit virus replication
and/or treat viral infection, or even serve as
models for new molecules. Numerous car-
rageenans and other polysaccarides have been
tested for their antiviral activity 4-7.
444
MONTANHA J.A., BOURGOUGNON N., BOUSTIE J. & AMOROS M.
We report herein our results concerning the
in vitro activity against HHV-1 and Poliovirus of
three carrageenan representatives of each struc-
tural type: λ- and ι–family (Gigartina acicu-
laris), ι-family (Euchema denticulatum) and κ
–family (Kappaphycus cottonii). Secondly, the
mechanism of viral inhibition was determined.
MATERIALS AND METHODS
Carrageenan
The carrageenans tested were extracted from
three marine red algae (Rhodophyceae) collect-
ed in different countries: Euchema denticulatum
(Philippines) Gigartina acicularis (Bretagne,
France) and Kappaphycus cottonii (Philippines).
They were extracted with hot distilled water, fil-
tered on diatomaceous earth and the filtrate was
poured into absolute ethanol, with stirring. The
precipitate was recovered and washed with 95°
ethanol, dehydrated with diethyl ether and dried
overnight at 50 °C 8. The chemical composition
of each extract differs in the content of sulfated
galactans and type of carrageenan. Each algae
seems to have different types of carrageenan:
Euchema denticulatum (ι-family) 9,10, Gigartina
acicularis (λ- and ι–family) 11 and Kappaphy-
cus cottonii (κ–family) 10.
The carrageenan samples were provided by
Prof. J.M. Kornprobst (SMAB/ISOMer – Nantes/
France). Stock solutions (10 mg/ml) were pre-
pared in PBS buffer and stored at -20 °C; and,
for the experiments, aliquots were diluted with
MEM to obtain the indicated concentrations.
Cells and viruses
African green monkey kidney cells (VERO
cell line ATCC CCL81) were grown in Eagle’s
minimum essential medium (MEM) supplement-
ed with 10 % newborn calf serum, 160 Units ml-
1 penicillin and 80 mg.ml–1 gentamicin. Cells
were routinely passed every three days. The
viruses used were the following: Human Hes-
pesvirus 1 (HHV-1 strain H29S) and poliovirus
type 2, a vaccinal strain Sabin II. Virus stocks
were propagated by serial passages on VERO
cells at a low multiplicity, incubated for 2 days,
then frozen and thawed three times. Afterwards
the preparation was cleaned by centrifugation at
low speed in order to remove the cell debris.
Virus titration was performed by the limiting di-
lution method 12. The virus titre was estimated
from cytopathogenicity and expressed as 50 %
tissue culture infectious doses ml–1 (TCID50/ml).
It was 2.0 x 105.5 TCID50/ml for HHV-1 and 2.0 x
106.57 TCID50 /ml for poliovirus.
Evaluation of cytotoxicity
Aiming assess the effect of carrageenan on
uninfected Vero cells, dilutions ranging from 6
mg/ml to 11.7 µg/ml (i.e. final concentrations in
wells from 1.5 mg/ml to 2.9 µg/ml) in the main-
tenance medium were added to VERO monolay-
ers (a 96-well microplate with 4.0 x 104 cells
per well). After incubation for 72 h, cytotoxicity
was determined by a microscopic examination
of the cell morphology. The concentration at
which the cell number was reduced to 50% in
relation to that of the controls was taken as the
50% cytotoxic concentration (CC50). The maxi-
mum tolerated concentration (MTC) was not
possible to determine because the carrageenan
solutions were very viscous. All assays were car-
ried out in triplicate.
Antiviral activity
Antiviral assays were carried out as described
previously 13. Dilutions of the carrageenan ex-
tracts ranging from 0.75 mg/ml to 0.003 µg/ml
were prepared in the maintenance medium and
added to confluent 1-day-old monolayers of
Vero cells. These cells were grown in microtitre
tissue culture plates just before inoculation, at a
multiplicity of infection (MOI) of 0.01 TCID50
per cell. Toxicity, cell, and virus controls were
run simultaneously. Plates were incubated at 37
°C for 72 h for HHV-1 or 32 h for poliovirus
(multiple replicative cycles). Then, the monolay-
ers were observed for cytopathic effect. The
concentration which inhibited 50% the viral cy-
topathic effect (compared to the virus control)
was expressed as the 50% inhibitory dose
(ID50). In order to quantify the antiviral activity,
the same plates were frozen and thawed three
times. The contents of the identical wells were
harvested, mixed and clarified by low-speed
centrifugation; and, virus titrations were per-
formed on the supernatant fluids by the limiting
dilution method 12. The antiviral activity of each
carrageenan extract was determined as the re-
duction factor (log10) of the viral titre by the
comparison with untreated controls. All experi-
ments were carried out three times.
Study of mode of action of the carrageenan
extracts against HHV-1 and Poliovirus
All experiments with carrageenans from Gi-
gartina acicularis (λ- and ι–family), Euchema
denticulatum (ι-family) and Kappaphycus cot-
tonii (κ–family) were performed at non toxic
concentration of 0.75 mg/ml.
445
Latin American Journal of Pharmacy - 28 (3) - 2009
Virus inactivation
Equal volumes of carrageenan extracts and
HHV-1 or Poliovirus stock suspension were
mixed and incubated for 1 h at 37 °C in order to
test possible virucidal activity. Thereafter, each
mixture was diluted 10-fold serially and infec-
tious titres were compared to those of controls 14.
Yield reduction after a single cycle of
replication
Vero cell monolayers cultured in 4-well cul-
ture plates were infected with HHV-1 or Po-
liovirus at a multiplicity of infections of about 1.
After 60 min at 37 °C, the unadsorbed viruses
were removed and the monolayers were
washed twice with MEM; subsequently, car-
rageenan extract dissolved in MEM was added.
After incubation of 18 h for HHV-1 or 8 h for
Poliovirus, the cultures were frozen and thawed
three times the cell debris was removed by low-
speed centrifugation. The supernatant virus
titres were determined by the limiting dilution
method and compared to that of the controls
without carrageenan extracts 14.
Culture pretreatment
Vero cell monolayers were pretreated with
each carrageenan extract for 24 h at 37 °C. After
being washed with MEM, the cells were ex-
posed to HHV-1 or Poliovirus at an Multiplicity
of Infections (MOI) of about 1. Following incu-
bation, the experiment was performed as de-
scribed above but no carrageenan extract was
added to the medium 14.
Study of viral DNA synthesis inhibition by
nucleic acid hybridization
Purification of HHV-1 DNA
Vero cells grown in 25 cm2tissue culture
flasks were infected at a high multiplicity (MOI
= 1), and HHV-1 adsorbed for 1 h at room tem-
perature, followed by washing with MEM to re-
move unadsorbed virus. From then on, only
maintenance medium (control) or carrageenan
extract dissolved in MEM (assays) was added.
The cultures were reincubated at 37 °C for 18 h.
The medium was discarded and the cells were
disrupted with 100 µl of lysis buffer. Subse-
quently, DNA was purified, as previously de-
scribed by Boom et al. 15, using a diatom sus-
pension. Samples were dissolved in 100 µl of
sterile water.
Hybridization probe
The digoxigenin-labelled DNA probe was a
non radioactive probe prepared by polymerase
chain reaction according to the method de-
scribed by Griffais et al. 16. The DNA template
was the U57 gene of HHV-1 obtained by a first
PCR 16. This template was amplified using two
primers: 5’-CTCACAGCCCCGAT-3’ and 5’-GTCC-
CGCGTTGC-3’. They were mixed with Thermus
aquaticus polymerase (Perkin Elmer Cetus, Nor-
walk, USA) and deoxynucleotide triphosphates:
d ATP, d CTP, d GTP and d UTP linked to
digoxigenin, dig-d UTP (Boehringer Mannheim,
Fr.). The solution was subjected to 38 cycles of
amplification in 3 steps: 92 °C, 55 °C and 72 °C.
Amplification was verified using agarose gel
electrophoresis.
Hybridization procedure
A twofold dilution of DNA extracts was per-
formed serially from 1/10 to 1/320 and dilutions
were heated up to 95 °C to denature DNA. The
samples were deposited on a Nylon membrane
HYBOND N (Amersham International Plc, UK)
and fixed to a HYBRI-SLOT manifold (BRL,
Maryland, USA). Afterwards, the filters were pre-
hybridized in thermally sealed plastic bags for 3
h at 42 °C in a prehybridization solution con-
taining 1% of a blocking agent (Boehringer
Mannheim, Fr.). After denaturation, a non-ra-
dioactive probe was added into the bag and hy-
bridization was carried out at 42 °C for 18 h,
with gentle shaking.
Detection of HHV-1 DNA probe binding
The hybrids were detected by enzyme-linked
immuno-assay using an antibody conjugate: an-
tidigoxigenin sheep antibody conjugated to al-
kaline phosphatase. The colour reaction was ini-
tiated at alkaline pH by the addition of 5-bro-
mo-4-chloro-3-indolyl phosphate (X-phosphate)
and nitroblue tetrazolium (NBT). (The detection
kit was purchased from Boehringer Mannheim,
Fr.). A blue precipitate indicated the presence of
HHV-1 DNA.
RESULTS AND DISCUSSION
We have investigated the anti-HHV-1 and the
antipoliovirus activity “in vitro” of three car-
rageenans extracts from three different algae
species collected in different countries.
The antiviral activity was first evaluated by
the inhibition of the cytopathic effect (CPE) in
cultures inoculated at a multiplicity of infection
of 0.01. The ID50 (mg/ml) values required for a
50% CPE inhibition and the selectivity indexes
(CC50/ID50) are indicated in Table 1.
446
MONTANHA J.A., BOURGOUGNON N., BOUSTIE J. & AMOROS M.
As shown in Table 1, the CC50 for the three
carrageenans was higher than 1.5 mg/ml (maxi-
mal concentration tested). Thus, the selectivity
index (ratio CC50/ID50) in the confluent cultures
used in antiviral assays range from 22 to more
than 545 for HHV-1; and from more than 6.6 to
more than 32 for Poliovirus at 0.01 MOI. Human
Herpes virus type I was more susceptible than
poliovirus to all carrageenan extracts. For HHV-
1 the highest selectivity index values were in the
same range as those previously found for
polysaccharides dextran sulfate and heparin
which were tested against HIV replication in
MT-4 and Molt-4 cell cultures 17 and for λ-car-
rageenan and pentosan polysulfate tested
against ASFV “in vitro” 18.
The effect on HHV-1 and Poliovirus replica-
tion was precisely quantified by infectious titre
reduction after several rounds of multiplication,
being the culture inoculated at 0.01 MOI. To be
considered active, a sample should induce at
least a 2 log10 decrease in virus titre in compari-
son with untreated virus control 19. According to
this criterion, all carrageenans tested were active
since the compounds reduced the virus titre be-
tween 2.0 and 3.0 log10 against HHV-1; whereas,
the activity against Poliovirus was lower with
titre reduction between 1.5 and 2.25 log10.
These results are shown in Table 1. The best re-
sults were obtained with carragenans from Gi-
gartina acicularis and Euchema denticulatum,
the more sulfated according to Amat 1.
Virus inactivation
No loss of infectivity was observed when
HHV-1 or Poliovirus was incubated for 1 h at 37
°C with Gigartina acicularis (λ- and ι–family),
Euchema denticulatum (ι-family) and Kappa-
phycus cottonii (κ–family) extracts of 0.75
mg/ml; it can be concluded that none of them
has a virucidal for the two viruses tested.
CPE inhibition (MOI 0.01) Yield reduction a(MOI 0.01)
HHV-1 Poliovirus HHV-1 Poliovirus
ID50 SI ID50 SI log10
Gigartina acicularis (λand τ) 0.00275 >545 0.047 >32 3.00 ± 0.86 2.50 ± 0.50
Euchema denticulatum (τ) 0.026 >71 0.115 >13 3.00 ± 0.25 1.50 ± 0.25
Kappaphycus cottonii (κ) 0.068 >22 0.227 >6.6 2.00 ± 0.25 1.50 ± 0.25
Table 1.In vitro antiviral activity of carrageenans against HHV-1 and Poliovirus. For three carrageenans the
CC50 is > 1.5 mg/ml; ID50 (mg/ml); SI (CC50/ID50). awhen compared with controls HHV-1 virus titre. Results
are presented as the mean ± SD from three independent tests.
Carrageenan
(type)
Effect of cell culture pretreatment
After a single round of replication, the HHV-
1 yield reduced 0.9, 0.9 and 0 log10 with Eu-
chema denticulatum (ι-family), Gigartina acicu-
laris (λ- and ι–family) and Kappaphycus cottonii
(κ–family), respectively. These results suggest
that the sulfated polysaccharides from Euchema
denticulatum and Gigartina acicularis have a
little effect on virus attachment or virus entry.
Effect of carrageenans on the yield of virus
from a single round of replication
In this experiment, Vero cells were infected
at a MOI of about 1; then, the carrageenans
were added to the culture medium 1 h posterior
to inoculation to exclude any effects on adsorp-
tion and penetration. After a single round of
replication, the HHV-1 yield reduced 2.0, 1.9
and 1.1 log10 with Euchema denticulatum (ι-
family), Gigartina acicularis λ- and ι–family)
and Kappaphycus cottonii (κ–family), respective-
ly. These results suggest that carrageenans from
Euchema denticulatum (ι-family) and Gigartina
acicularis exert their antiviral effect, in part, by
the inhibition of the virus attachment and the in-
terference in a subsequent stage of the virus at-
tachment. To determine this step, HHV-1 DNA
synthesis was evaluated by a hybridization tech-
nique in cultures where the most active car-
rageenans (Euchema denticulatum - ι-family
and Gigartina acicularis - λ- and ι–family) were
added. The effect of carrageenans on HHV-1
DNA synthesis was measured by nucleic acid
hybridization with a digoxigenin-labelled HHV-1
DNA probe. The DNA of untreated virus-infect-
ed controls, serially diluted twofold, was de-
tectable up to a 1:320 dilution whereas the DNA
of uninfected untreated control cells could not
be detected even at a 1:10 dilution, which
proved the probe specificity. In cultures treated
with carrageenans from Euchema denticulatum
447
Latin American Journal of Pharmacy - 28 (3) - 2009
and Gigartina acicularis, DNA synthesis was re-
duced threefold and twofold, respectively, with
0.75 mg/ml (Fig. 1). These results indicate that
carrageenans interfere with viral DNA synthesis.
This test did not, however, allow us to conclude
that replication was actually disrupted. Inhibi-
tion could result from another phenomenon oc-
curring upstream, such as DNA polymerase in-
terference.
These results are consistent with the yield re-
duction after a single cycle of replication, al-
though the decrease of DNA synthesis alone
could not explain the reduction of virus titre.
Other events of the replication cycle, such as
synthesis of late proteins or assembly step can
be inhibited by these polysaccharides. These
findings suggest that these sulfated polysaccha-
rides inhibit a subsequent step in virus replica-
tion to viral internalization but prior to the onset
of late viral protein synthesis as described by
Gonzáles et al. 20 for other polysaccharides
(Table 2).
If we consider the structure-activity relation-
Yield reduction log10 MOI = 1
Culture pretreatment After a single cycle of replication
HHV-1 Poliovirus HHV-1 Poliovirus
Gigartina acicularis (λand τ) 0.9 ± 0.25 0.3 ± 0.10 1.9 ± 0.25 0.5± 0.25
Euchema denticulatum (τ) 0.9 ± 0.25 0.5 ± 0.20 2.0 ± 0.50 0.5 ± 0.10
Kappaphycus cottonii (κ) 0 0.2 ± 0.25 1.1 ± 0.50 0.3 ± 0.10
Table 2. Mechanism of action of the carrageenans. Results are presented as the mean ± SD from three indepen-
dent tests.
Carrageenan
(type)
Figure 1. Slot hybridization of DNA extracts from
HHV-1 infected cells untreated (A1 to A6) and treated
with carrageenans (A7 to A12, B1 to B6 and B7 to
B12).A1 to A6: DNA of untreated virus control (HHV-
1). A7 to A12: DNA of cultures treated with 0.37
mg/ml of Euchema denticulatum (τ-family). B1 to
B6: DNA of cultures treated with 0.75 mg/ml of Eu-
chema denticulatum (τ-family). B7 to B12: DNA of
cultures treated with 0.75 mg/ml of Gigartina acicu-
laris (λ- and τ–family).
ships, the more actives carrageenans (Gigartina
acicularis -λ- and ι–family and Euchema den-
ticulatum -ι-family) are more sulfated than the
less active ones (Kappaphycus cottonii, κ–fami-
ly). These data were also observed by Gonzáles
et al. 20. Regarding the spectrum the of action of
the carrageenans tested, they were more active
against HHV-1 (enveloped virus) and less active
against poliovirus (a naked virus). These results
are also in agreement with the results obtained
by Gonzáles et al. 20.
We conclude that these extracts have an im-
portant antiviral activity and are potential candi-
dates for further studies of mechanism of action.
Acknowledgments. The present paper is dedicated
to the memory of Professor Loïc Girre. We are grate-
ful to Professor J. M. Kornprobst (SMAB/ISOMer –
Nantes/France) for providing the carrageenan sam-
ples. We also thank Mrs. M. Garel and Mrs. F. Baril
for technical assistance. A Research Fellowship held
from CAPES, Brazil was received by J.A. Montanha.
REFERENCES
1. Amat, M.A. (1989) Océanis. 15: 661-71.
2. Boisson-Vidal, C., F. Haroun, M. Ellouali, C.
Blodin, A.M. Fischer, A. de Agostin & J. Joze-
fonvicz (1995) Drug Future 20: 1237-49.
3. Vlietinck, A.J., T. De Bruyne & D.A. Vanden
Berghe (1997) Curr. Org. Chem. 1: 307-44.
4. Adhikari, U., C.G. Mateu , K. Chattopadhyay,
C.A. Pujol, E B. Damonte & B. Ray (2006) Phy-
tochemistry 67: 2474–82.
5. Cáceres, P.J., M.J. Carlucci, E.B. Damonte, B.
Matsuhiro & E.A. Zúñiga ( 2000) Phytochem-
istry 53: 81-6.
6. Carlucci, M.J., M. Ciancia, M.C. Matulewicz,
A.S. Cerezo & E.B. Damonte(1999) Antivir.
Res. 43: 93-102.
7. Carlucci, M.J., L.A. Scolaro, M.D. Noseda, A.S.
Cerezo & E.B. Damonte (2004) Antivir. Res.
64: 137-41.
8. Bourgougnon, N., M. Lahaye, J-C. Chermann &
448
MONTANHA J.A., BOURGOUGNON N., BOUSTIE J. & AMOROS M.
J-M. Kornprobst (1993) Bioorg. Med. Chem.
Lett. 3: 1141-6.
9. Fostier, A.H. (1989) Contribuition à la valori-
sation d’algues des Côtes Sénégalaises produc-
trices de iota carraghénanes. [Thèse de Doc-
torat, mention Océanologie]. Perpignan, France.
10. Bellion, C., G., Brigard, J.-C. Prome, D. Welti
& S. Bocik (1983) Carbohyd. Res. 119: 31-48.
11. Bourgougnon, N. (1991) Les Carraghénanes.
Essais de séparation & evaluation pharma-
cologique. [Diplôme d’Etudes Approfondies
d’Océanologie biologie Connaissances dês
Producteurs primaire]. Université de Nantes.
Nantes, France.
12. Payment, P. & M. Trudel, eds. ( 1989) “Manuel
de techniques virologiques”, Presses de l’Uni-
versité du Québec, Québec, pp. 39-40.
13. Fritz, D., C.R.Venturi, S. Cargnin, J. Schripse-
ma, P.M. Roehe, J.A. Montanha & G.L.von Pos-
er (2007) J. Ethnopharmacol. 113: 517-20.
14. Montanha, J.A., M. Amoros, J. Boustie & L.
Girre (1995) Planta Med. 61: 419-24.
15. Boom, R., C.J.A. Sol, M.M.M. Salimans, C.L.
Jansen, P.M.E. Wertheim-Van Dillen & J.Van
Der Noordaa (1990) J. Clin. Microbiol. 28: 495-
503.
16. Griffais, R., P.M. Andre & M. Thibon (1990)
Res. Virology. 141: 331-5.
17. Ito, M., M Baba, A. Sato, R. Pauwels, E. De
Clercq & S. Shigeta (1987) Antivir. Res. 7: 361-
7.
18. García-Villalón, D. & C. Gil-Fernández (1991)
Antivir. Res. 15: 139-48.
19. Van den Berghe, D.A., A.J. Vlietinck & L. Van
Hoof (1986) B.I. Pasteur 84: 101-47.
20. Gonzáles, M.E., B. Alarcón & L. (Carrasco
1987) Antimicrob. Agents Ch. 31: 1388-93.
