Ungeremine and its hemisynthesized analogues as bactericides against flavobacterium columnare

Article abstract of DOI:10.1021/jf304586j

The Gram-negative bacterium Flavobacterium columnare is the cause of columnaris disease, which can occur in channel catfish (Ictalurus punctatus). In a previous study, the betaine-type alkaloid ungeremine, 1, obtained from Pancratium maritimum L. was found to have strong antibacterial activity against F. columnare. In this study, analogues of 1 were evaluated using a rapid bioassay for activity against F. columnare to determine if the analogues might provide greater antibacterial activity and to determine structure-activity relationships of the test compounds. Several ungeremine analogues were prepared by hydrochlorination of the alkaloid and by selenium dioxide oxidation of both lycorine, 7, and pseudolycorine, 8, which yielded the isomer of ungeremine, 3, and zefbetaine, 4, respectively. The treatment of lycorine with phosphorus oxychloride allowed the synthesis of an anhydrolycorine lactam, 5, showing, with respect to 1, the deoxygenation and oxygenation of C-2 and C-7 of the C and B rings, respectively. The results of the structure-activity relationship studies showed that the aromatization of the C ring and the oxidation to an azomethine group of C-7 of the B ring are structural features important for antibacterial activity. In addition, the position of the oxygenation of the C ring as well as the presence of the 1,3-dioxole ring joined to the A ring of the pyrrolo[de]phenanthridine skeleton also plays a significant role in imparting antibacterial activity. On the basis of 24-h 50% inhibition concentration (IC₅₀) results, ungeremine hydrochloride, 2, was similar in toxicity to 1, whereas 5 had the lowest activity. Analogue 2 is soluble in water, which may provide the benefit for use as an effective feed additive or therapeutant compared to ungeremine.

Full text of DOI:10.1021/jf304586j

Article
pubs.acs.org/JAFC
Ungeremine and Its Hemisynthesized Analogues as Bactericides
against Flavobacterium columnare
Kevin K. Schrader,*,† Fabiana Avolio,‡ Anna Andolfi,‡ Alessio Cimmino,‡ and Antonio Evidente‡,§
†Natural Products Utilization Research Unit, Thad Cochran National Center for Natural Products Research, Agricultural Research
Service, U.S. Department of Agriculture, P.O. Box 8048, University of Mississippi, University, Mississippi 38677, United States
‡Dipartimento di Scienze Chimiche, Complesso Universitario Monte Sant’Angelo, Universita
̀
di Napoli Federico II, Via Cintia 4,
80126 Napoli, Italy
§Istituto di Chimica Biomolecolare, CNR, Pozzuoli, Italy
ABSTRACT: The Gram-negative bacterium Flavobacterium columnare is the cause of columnaris disease, which can occur in
channel catfish (Ictalurus punctatus). In a previous study, the betaine-type alkaloid ungeremine, 1, obtained from Pancratium
maritimum L. was found to have strong antibacterial activity against F. columnare. In this study, analogues of 1 were evaluated
using a rapid bioassay for activity against F. columnare to determine if the analogues might provide greater antibacterial activity
and to determine structure−activity relationships of the test compounds. Several ungeremine analogues were prepared by
hydrochlorination of the alkaloid and by selenium dioxide oxidation of both lycorine, 7, and pseudolycorine, 8, which yielded the
isomer of ungeremine, 3, and zefbetaine, 4, respectively. The treatment of lycorine with phosphorus oxychloride allowed the
synthesis of an anhydrolycorine lactam, 5, showing, with respect to 1, the deoxygenation and oxygenation of C-2 and C-7 of the
C and B rings, respectively. The results of the structure−activity relationship studies showed that the aromatization of the C ring
and the oxidation to an azomethine group of C-7 of the B ring are structural features important for antibacterial activity. In
addition, the position of the oxygenation of the C ring as well as the presence of the 1,3-dioxole ring joined to the A ring of the
pyrrolo[de]phenanthridine skeleton also plays a significant role in imparting antibacterial activity. On the basis of 24-h 50%
inhibition concentration (IC50) results, ungeremine hydrochloride, 2, was similar in toxicity to 1, whereas 5 had the lowest
activity. Analogue 2 is soluble in water, which may provide the benefit for use as an effective feed additive or therapeutant
compared to ungeremine.
KEYWORDS: alkaloids, Amaryllidaceae, bactericides, channel catfish, Flavobacterium columnare, Pancratium maritimum, ungeremine
feeds in agriculture, which make the future use of medicated
feed in catfish aquaculture uncertain.
INTRODUCTION
■
Farm-raised channel catfish (Ictalurus punctatus) currently
comprise the largest segment of the freshwater aquaculture
industry in the United States, and Mississippi supplies over half
of the farm-raised catfish processed annually. The two most
common bacterial diseases of pond-raised catfish are colum-
naris disease and enteric septicemia of catfish (ESC), which
cause large economic losses to producers of pond-raised
catfish.1 Columnaris disease and ESC are caused by the Gram-
negative bacteria Flavobacterium columnare and Edwardsiella
ictaluri, respectively.1 Although both of these bacterial
pathogens can be the cause of primary infections, columnaris
disease can also develop as a secondary pathogen, becoming
acute with rapid mortality at rates as high as 50−60% in pond
populations.1
The most common management approaches used by
producers to control F. columnare and E. ictaluri are the use
of medicated feeds, attenuated vaccines, and nonantibiotic
therapeutants. The U.S. Food and Drug Administration has
approved Aquaflor (50% w/w florfenicol active ingredient) for
treating ESC and columnaris disease, currently under condi-
tional approval, in channel catfish.2 However, the use of
antibiotics has several concerns such as the potential develop-
ment of antibiotic-resistant strains of E. ictaluri and F. columnare
and the environmental impact from the use of antibiotic-laden
Plants remain a vastly unexplored source for antibacterial
compounds, especially in the discovery of novel antibacterial
compounds as alternatives to commercially available antibiotics
currently used in aquaculture (e.g., Aquaflor). Previous research
discovered that ungeremine had strong antibacterial activities
against two isolates of F. columnare (ALM-00-173 and
BioMed).3 Ungeremine, 1 (Figure 1), is a betaine-type alkaloid
isolated from Pancratium maritimum L. (family Amaryllida-
ceae),4 and it can also be isolated from other species of
Amaryllidaceae including Ungernia minor, Crinum americanum,
Crinum asiaticum, and Zephyranthes flava.4 In addition,
ungeremine has been obtained as the main metabolite of the
microbial degradation of lycorine, 7 (Figure 1), the main
Amaryllidaceae alkaloid,5 by an unidentified species of the
bacterium Pseudomonas (strain ITM 311) that was isolated
from the rhizosphere of Sternbergia lutea Ker Gawl.6 This plant
species is also the best source of lycorine.7 Ungeremine was
also obtained by selenium dioxide oxidation of lycorine by
applying a modified protocol6 with respect to those previously
Received: October 26, 2012
Revised: January 17, 2013
Accepted: January 20, 2013
Published: January 20, 2013
© 2013 American Chemical Society
1179
dx.doi.org/10.1021/jf304586j | J. Agric. Food Chem. 2013, 61, 1179−1183

Journal of Agricultural and Food Chemistry
Article
M. Fales, Department of Health, Education, and Welfare, Bethesda,
MD, USA.
Plant Materials. Fresh bulbs of P. maritimum L. were collected
from sandy hills on the northern coastal strip of Egypt (Baltim) during
the flowering and fruit-producing stage. The plant was kindly identified
by Professor Dr. N. El Hadidy, University of Cairo, Egypt. A voucher
sample is deposited in the Department of Pharmacognosy, Faculty of
Pharmacy, University of Alexandria, Egypt. The bulbs were dried by
ventilation in an oven at 35 °C, minced, and stored at −20 °C. The
plant S. lutea Ker Gawl, collected near Bari, Italy, during the
weathering period, was identified by Prof. Arrigoni, Dipartimento di
̀
Biologia e Patologia Vegetale, Universita di Bari, Italy, where a voucher
specimen has been deposited. The fresh bulbs were dried, minced, and
stored as reported for P. maritimum.
Ungeremine, 1. Initially, 1 was extracted with EtOH from the dried
and minced bulbs of P. maritimum L. (1.6 kg) as previously reported.4
The crude extract was submitted to different steps of extraction
followed by chromatography as previously reported4 to give 1 (15 mg)
as brilliant orange needles, which were recrystallized from MeOH/
C6H6; mp was 267−271 °C, which is very close to the reference mp of
268−270 °C.4 Compound 1 was also obtained by oxidation of lycorine
using a modified procedure6 iwith respect to that previously reported.8
Briefly, a suspension of lycorine (375 mg) and sublimed SeO2 (250
mg) in EtOH (3 mL) was warmed on a steam bath to effect solution.
After the precipitation of selenium metal, the reaction was heated
under reflux for 4 h and filtered while hot. The partial evaporation of
ethanol under vacuum left a yellow solid. The solution was dissolved in
water and treated with 4 N NaOH to give yellow flocculi. The crude
betaine was collected by centrifugation (10000 rpm for 45 min).
Crystallization from EtOH yielded light orange needles of 1 (76 mg):
IR vmax 1649, 1615, 1592, 1449, 1410, 1013 cm−1; UV λmax nm (log ε)
369 (3.40), 287 (3.75), 261 (4.06); ESI MS (+) m/z 266 [M + H]+.
The 1H and 13C NMR data are very similar to those previously
reported.4,6
Figure 1. Chemical structures of ungeremine, 1, lycorine, 7, and
respective analogues.
Ungeremine Hydrochloride, 2. Compound 1 (20 mg) dissolved in
water was treated with 1 N HCl (2 mL) at room temperature. A pale
yellow precipitate was immediately generated and collected by
centrifugation (10000 rpm × 30 min). The solid was dried as
previously reported8 at 137 °C to give derivative 2 as yellow needles
(15 mg). Elemental analysis: C, 63.62; H, 4.06; Cl, 11.72; N, 4.60; O,
15.93 [calcd for C16H12ClNO3, C, 63.69; H, 4.01; Cl, 11.75; N, 4.64;
O, 15.91 (Fales et al.8 found C16H11NO3·HCl·H2O, C, 59.69; H, 4.33;
N, 5.28)]. ESI MS (+) m/z 266 [M + H]+.
reported in a study on the actions of oxidizing agents on this
alkaloid.8
While considering the economics of obtaining a sufficient
amount of ungeremine by the oxidation of lycorine, previously
isolated in large amounts from S. lutea,7 and the possibilities of
preparing ungeremine analogues from lycorine and other
Amaryllidaceae alkaloids by using different oxidizing agents,
we developed a structure−activity relationship (SAR) study.
This study was performed in conjunction with the study to
evaluate natural or hemisynthetic ungeremine analogues for
potentially increased antibacterial activities and/or specificities
against F. columnare. The results of the bioassay evaluation of
the ungeremine analogues and the SAR study are reported here.
Ungeremine Isomer, 3. According to a modified procedure with
respect to that previously reported,8 lycorine (250 mg) was dissolved
in warm 10% of AcOH (10 mL), and then Hg(AcO)2 (700 mg) was
added. The solution was heated on a steam bath for 3 min, and a dense
precipitate of Hg(OAc)2 was formed. After 45 min, the reaction
filtered hot and the 10% HOAc was removed by vacuum evaporation.
The residue (676 mg) was purified by column RP-18 Lichroprep
column, using as solvent H2O/CH3CN (8:2, v/v), giving three groups
of homogeneous fraction. The residue (13.3 mg) of the second
fraction showed the presence of two compounds, which were purified
by preparative TLC on silica gel using as solvent BuOH/H2O/AcOH
(3:1:1, v/v/v), subsequently yielding 1 (9.7 mg, Rf 0.49) and its
isomer, 3 (2.7 mg, Rf 0.42). Compound 3 had the following elemental
analysis: C, 72.51; H, 4.23; N, 5.21; O, 18.13 (calcd for C16H11NO3, C,
72.45; H, 4.18; N, 5.28; O, 18.09). ESI MS (+) m/z 266 [M + H]+.
Zefbetaine, 4. According to a modified procedure with respect to
that previously reported,4 a suspension of pseudolycorine, 8 (15 mg),
and sublimed SeO2 (9 mg) in EtOH (1 mL) was warmed on a steam
bath to effect solution. After the precipitation of selenium metal, the
reaction was heated under reflux for 4 h and filtered while hot. The
solid residue (20 mg) left by evaporation of the ethanol was purified
by preparative TLC on silica gel using as solvent CHCl3/MeOH/
EtOAc (2:2:1, v/v/v) giving 4 as an amorphous yellow solid (5 mg, Rf
0.47): IR vmax 2984, 1652, 1612, 1576, 1452, 1410, 1013 cm−1; UV
λmax nm (log ε) 411 (3.21), 265 (3.59), 220 (4.00); ESI MS (+) m/z
MATERIALS AND METHODS
■
General Experimental Procedures. Melting point (mp) was
measured with an Axioskop Zeiss (Oberkochen, Germany); elemental
analyses were performed with a Perkin-Elmer (Norwalk, CT, USA)
AA Analyst 700. IR spectra were recorded as glassy film on a Perkin-
Elmer Spectrum One FT-IR spectrometer, and UV spectra were taken
in MeCN solution on a Perkin-Elmer Lambda 25 UV−vis
1
spectrophotometer. H and 13C NMR spectra were recorded at 600
and 125 MHz, respectively, in CDCl3, on Bruker (Kalsruhe, Germany)
spectrometers. The same solvent was used as internal standard. ESI
MS spectra were recorded on an Agilent Technologies (Milan, Italy)
6120 quadrupole LC-MS. Analytical and preparative thin layer
chromatography (TLC) were performed on silica gel plates (0.25
mm Kieselgel 60 and 0.5 mm F254; Merck, Darmstadt, Germany); the
spots were visualized by exposure to UV light or by exposure to I2
vapors. Column chromatography was performed on 0.063−0.200 mm
silica gel (Merck, Rahway, NJ, USA) or 40−63 μm, 10 × 240 mm,
Lichroprep RP18 (Merck, Rahway, NJ, USA) working at a medium
pressure of 3 bar. Pseudolycorine was generously supplied by Prof. H.
1
268 [M + H]+. The H and 13C NMR data are very similar to those
previously reported.4
1180
dx.doi.org/10.1021/jf304586j | J. Agric. Food Chem. 2013, 61, 1179−1183

Journal of Agricultural and Food Chemistry
Article
Anhydrolycorine Lactam, 5. According to a previously reported
procedure,9 a mixture of 200 mg of lycorine and 680 mg of POCl3
(414 μL) was refluxed at 100 °C for 15 min. The mixture was poured
dropwise into ice-cold water (1 mL) to provide a complete solution.
This solution was basified with 10% NaOH (50 μL) and 10% Na2CO3
(75 μL) and then extracted with ethyl ether (3 × 3 mL). The
combined ether extracts were dried (Na2SO4), filtered, and evaporated
under reduced pressure. The residue was dissolved in MeOH (2 mL),
and then 10% KOH (200 mL) was added. After reflux for 1 h, the
reaction mixture was diluted with H2O (204 μL) and extracted (3 × 5
mL) with a solution of C6H6/ethyl ether (10:3, v/v). The organic
extracts were combined, dried (Na2SO4), filtered, and evaporated
under reduced pressure. The residue (33 mg) was purified by
preparative TLC on silica gel using as solvent solvent CHCl3/MeOH/
EtOAc (2:2:1, v/v/v), yielding the derivative 5 (4.3 mg) as an
RESULTS AND DISCUSSION
■
Ungeremine, 1 (Figure 1), was obtained from bulbs of P.
maritimum L. and also by selenium dioxide oxidation of
lycorine as previously reported.4,6 Lycorine, 7 (Figure 1), was
obtained as white prisms in large amount from bulbs of S. lutea
as previously reported.7 Compound 1 was converted into its
hydrochloride, 2 (Figure 1), by treatment with HCl, whereas its
isomer, 3 (Figure 1), was obtained by oxidation of 7 with
Hg(AcO)2 by applying a modified protocol with respect to that
previously reported.8 Zefbetaine, 4 (Figure 1), another betaine-
type alkaloid previously isolated from P. maritimum L.4 and also
from Z. flava,13 was obtained by selenium dioxide oxidation of
pseudolycorine, 8 (Figure 1), another Amaryllidaceae alkaloid5,
by applying a modified protocol with respect to that previously
reported.4 The physical and spectroscopic data of 1, its
analogues 2 and 3, and 4 were in full agreement with those
previously reported4,6,8 and confirmed by ESI MS as reported
in detail in the corresponding experimental paragraphs.
Compound 2 differs from 1 only by the presence of a
quaternary protonated nitrogen atom that, depending on the
pH, could be reversibly reconverted into the nonprotonated
form. The isomer 3 differs from 1 substantially by the
oxygenation of the aromatized C ring, the negatively charged
oxygen being at C-1 in 3 instead of at C-2 as in 1. Finally, 4
differs from 1 only by the reductive opening of the dioxole ring
joined to the A ring, which in 4 was converted so that the
hydroxy and methoxy groups were bonded to C-9 and C-10,
respectively. Finally, in the attempt to prepare anhydrolycorine,
6 (Figure 1), by reaction of 7 with POCl3, anhydrolycorine
lactam, 5 (Figure 1) was obtained. Its physical and
spectroscopic data were in full agreement with those previously
reported.6,10,14
amorphous solid: IR νmax 1630, 1610, cm−1 [Hansel and Thober10
̈
found IR (KBr) 3045, 2975, 2915, 2780, 1640, 1620, 1585, 1505,
1495, 1470, 900−700 cm−1]; UV (EtOH) λmax nm (log ε) 341 (3.88),
1
324 (3.88), 309 (3.88), 273 (4.24), 246 (4.54), 242 (4.58). The H
data are very similar to those previously reported.6 Elemental analysis:
C, 72.23; H, 4.60; N, 5.19; O, 17.94 (calcd for C16H12NO3, C, 72.17;
H, 4.54; N, 5.26; O, 18.03). ESI MS (+) m/z 266 [M + H]+.
Lycorine, 7. Lycorine was obtained by extraction with 1% H2SO4
from dried bulbs of S. lutea. The free base obtained by alkalinization of
the acid extracts was crystallized as white prisms as detailed previously
reported (11 g/kg of dried plant bulbs).7
Bactericide Bioassay. Two isolates of F. columnare [ALM-00-173
(genomovar II) and BioMed (genomovar I)] were obtained from Dr.
Covadonga Arias (Department of Fisheries and Allied Aquacultures,
Auburn University, AL, USA). Purity of cultures of the F. columnare
isolate was assured by streaking the bacteria for isolation onto
modified Shieh (MS) agar plates (pH 7.2−7.4)11 and then checking
after incubation at 29 1 °C for 3−5 days that only one bacterial
colony type was present. Prior to conducting the bioassay, we used
single colonies of the test cultures to prepare the assay culture
materials by culturing each isolate in 75 mL of modified Shieh broth
(18 h for BioMed and 24-h for ALM-00-173) at 29 1 °C at 150 rpm
on a rotary shaker. Test compounds were evaluated for antibacterial
activity using a rapid 96-well microplate bioassay12 of which
procedures are briefly described below. Florfenicol and oxytetracycline
HCl (Sigma-Aldrich, St. Louis, MO, USA), antibiotics that are utilized
in medicated feed, were included as positive drug controls for each
bioassay. In addition, control wells (no test compound or solvent
added) and wells containing ungeremine were included in each assay.
Technical grade methanol was used to dissolve the test compounds,
except that technical grade methylene chloride was used to dissolve 5.
Final concentrations of test compounds and each drug control in the
microplate wells were 0.01, 0.1, 1.0, 10.0, 100.0, and 1000.0 μM, and
three replications were used for each dilution of each test compound
and controls. To determine the 24-h 50% inhibition concentration
(IC50) and minimum inhibition concentration (MIC) of each test
compound, sterile 96-well polystyrene microplates (Corning Costar
Corp., Acton, MA, USA) with flat-bottom wells were used to conduct
the bioassay. Dissolved test compounds were added to microplate
wells (10 μL/well), and the solvent was allowed to completely
evaporate at room temperature before 0.5 MacFarland bacterial
culture12 was added to the microplate wells (200 μL/well).
Compound 1 and its hemisynthetic analogues 2−5 were
assayed against F. columnare [isolate ALM-00-173 (genomovar
II) and isolate BioMed (genomovar I)] as reported in detail
under Materials and Methods, and the results are reported in
Table 1. The strong bactericidal activities of 1 previously
observed3 were confirmed again. The 24-h IC50 results of 1 for
F. columnare ALM-00-173 and F. columnare BioMed were 0.54
0.06 and 2.15 0.05 μM, respectively. Genomovar II isolates
such as F. columnare ALM-00-173 are more pathogenic for
channel catfish.15 For both isolates, the MIC of 1 was 1.0
μM. As expected, 2 showed similar bactericidal activity as 1.
The 24-h IC50 results of 2 for F. columnare ALM-00-173 and F.
columnare BioMed were 1.8 0 and 3.0 0.1 μM, respectively,
0
whereas the MIC results were 1.0
0 and 10.0
0 μM,
respectively. The 24-h IC50 relative-to-drug control (RDC)
values were slightly higher for 2 compared to 1. It is believed
that 2 in solution resulted in conversion of the nitrogen atom
into the nonprotonated form, which would thereby result in 2
being converted into 1. These results agreed with those
previously observed when 7 and its hydrochloride salt were
assayed for their ability to inhibit the ascorbic acid biosynthesis
in potato tubers16 and when they were assayed more recently
for their anticancer activity on different tumor cell lines.17
The isomer 3 was significantly less active than 1 against both
isolates of F. columnare, and this could substantially depend on
the different position of the negatively charged oxygen atom on
the C ring. In fact, it was bonded at C-2 in 1 and at C-1 in 3.
The 24-h IC50 results of 3 for F. columnare ALM-00-173 and F.
Microplates were maintained at 29
1 °C in an incubator. A
SpectraCount microplate photometer (Packard Instrument Co.,
Meriden, CT, USA) was used to measure the absorbance (630 nm)
of the microplate wells at time 0 and 24-h. The means and standard
errors of absorbance measurements were calculated, graphed, and
compared to controls to determine the 24-h IC50 and MIC for each
test compound.12 The 24-h IC50 and MIC results for each test
compound were divided by the respective 24-h IC50 and MIC results
obtained for the positive controls florfenicol and oxytetracycline to
determine the relative-to-drug-control florfenicol (RDCF) and
relative-to-drug-control oxytetracycline (RDCO) values.
columnare BioMed were 14.6
11.4 and 31.0
0 μM,
1181
dx.doi.org/10.1021/jf304586j | J. Agric. Food Chem. 2013, 61, 1179−1183

Journal of Agricultural and Food Chemistry
Article
respectively, and MIC results were 10.0 0 and 100.0 0 μM,
respectively (Table 1).
Compounds 4 and 5 were among the least active ungeremine
analogues. For F. columnare ALM-00-173 and F. columnare
BioMed, the 24-h IC50 values of 4 were 45.5 4.5 and 15.0
2.0 μM, respectively, whereas the MIC values of 4 were 100.0
0 and 10.0 0 μM, respectively (Table 1). On the basis of the
24-h IC50 results, 5 was less active against F. columnare BioMed
than against F. columnare ALM-00-173, with 24-h IC50 values of
195.0
5.0 and 37.0
13.0 μM, respectively. A previous
study18 determined that the 24-h IC50 values of 7 for F.
columnare ALM-00-173 and F. columnare BioMed were 95.0
0 and 190 0 μM, respectively, whereas the MIC value for
both isolates was 100.0 0 μM (Table 1). Therefore, 7 was
determined to also be among the least active compounds.
These results showed that the 1,3-dioxole ring, lacking in 4 as
consequence of its reductive opening, plays a significant role in
providing bactericidal activity. The reduced activity of derivative
5 was due to the oxidation of H2C-7 to the carbonyl group
generating the corresponding tertiary amide, which could
probably be converted into the corresponding enol form under
acidic conditions. The modifications of 5 determine the loss of
the betaine nature present in 1, which is a fundamental
structural feature for the activity. These results are in
disagreement with the results of previous structure−activity
relationship studies such as the testing of antibiotic activity
against Corynebacterium fascians6 and the study on the
inhibition of ascorbic acid biosynthesis in potato tubers using
lycorine and several hemisynthetic derivatives and analogues,
which demonstrated that the different oxidation state of C-7 did
not affect the activity.19 Instead, the lactam nature of C-7 as in 5
determined the strong reduction of activity when it was assayed
in comparison with lycorine against Plasmodium falciparum.14
This information may prove useful in future mode-of-action
studies for 1 as related to its antibacterial activity against F.
columnare.
Overall, none of the ungeremine analogues were more active
than 1 against either isolate of F. columnare. However, 2 was
nearly as active as 1 against both isolates on the basis of 24-h
IC50 results; 2 was less active than 1 against F. columnare
BioMed, whereas 2 had the same activity as 1 against F.
columnare ALM-00-173 on the basis of MIC results. Analogue 2
does provide water solubility to the parent compound
ungeremine, which is an important consideration in terms of
an adequate delivery mechanism (e.g., feed additive or
therapeutant) to test fish during any future efficacy studies.
In conclusion, the results of the SAR studies showed that the
aromatization of the C ring as well as the oxidation to
azomethine group of C-7 of the B ring, which determines the
betaine nature of ungeremine, are structural features important
for the bactericidal activity. The position of the oxygenation of
the C ring as well as the presence of the 1,3-dioxole ring joined
to the A ring of the pyrrolo[de]phenthantridine skeleton also
plays a significant role in providing antibacterial activity.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 1 (662) 915-1144. Fax: 1 (662) 915-1035. E-mail:
kevin.schrader@ars.usda.gov.
Funding
This work was supported in part by a grant from the Italian
Ministry of University and Research.
1182
dx.doi.org/10.1021/jf304586j | J. Agric. Food Chem. 2013, 61, 1179−1183

Journal of Agricultural and Food Chemistry
Article
microbial and plant metabolites as potential products for pest
management. Chem. Biodiver. 2010, 7, 2261−2280.
(19) Evidente, A.; Arrigoni, O.; Liso, R.; Calabrese, G.; Randazzo, G.
Further experiments on structure-activity relationships among the
lycorine alkalods. Phytochemistry 1986, 25, 2739−2743.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge Dominique Melck for recording the
NMR spectra in the laboratory of the Istituto di Chimica
Biomolecolare del CNR, Pozzuoli, Italy. We greatly appreciate
the technical assistance of Phaedra Page and Marcuslene
Harries.
REFERENCES
■
(1) Plumb, J. A.; Hanson, L. A. Health Maintenance and Principal
Microbial Diseases of Cultured Fishes, 3rd ed.; Wiley-Blackwell: Ames,
IA, 2011.
(2) Gaunt, P. S.; Gao, D.; Sun, F.; Endris, R. G. Efficacy of florfenicol
for control of mortality caused by Flavobacterium columnare infection
in channel catfish. J. Aquat. Anim. Health 2010, 22, 115−122.
(3) Schrader, K. K.; Andolfi, A.; Cantrell, C. L.; Cimmino, A.; Duke,
S. O.; Osbrink, W.; Wedge, D. E.; Evidente, A. A survey of phytotoxic
microbial and plant metaboites as potential natural products for pest
management. Chem. Biodiver. 2010, 7, 2261−2280.
(4) Abou-Donia, A. H.; Abib, A.-A.; El Din, S. A.; Evidente, A.;
Gaber, M.; Scopa, A. Two betaine-type alkaloids from Egyptian
Pancratium maritimum. Phytochemistry 1992, 31, 2139−2141.
(5) Kornienko, A.; Evidente, A. Chemistry, biology, and medicinal
potential of narciclasine and its congeners. Chem. Rev. 2008, 108,
1982−2014.
(6) Evidente, A.; Randazzo, G.; Surico, G.; Lavermicocca, P.;
Arrigoni, O. Degradation of lycorine by Pseudomonas species strain
ITM 311. J. Nat. Prod. 1985, 48, 564−570.
(7) Evidente, A.; Iasiello, I.; Randazzo, G. An improved method for
the large-scale preparation of lycorine. Chem. Ind. 1984, 348−349.
(8) Fales, H. M.; Warnhoff, E. W.; Wildman, W. C. Alkaloids of the
Amaryllidaceae. VI. The action of oxidizing agents on lycorine and
caranine. J. Am. Chem. Soc. 1955, 77, 5855−5890.
(9) Takeda, K.; Kotera, K.; Mizukami, S. Partial synthesis of caranine.
J. Am. Chem. Soc. 1958, 80, 2562−2567.
(10) Hansel, R.; Thober, H. Anhydrolycorinon, ein neues Alkaloid
̈
aus Amaryllis belladonna L. Arch. Pharm. 1982, 315, 767−768.
(11) Decostere, A.; Ducatelle, R.; Haesebrouck, F. Flavobacterium
columnare (Flexibacter columnaris) associated with severe gill necrosis
in koi carp (Cyprinus carpio L.). Vet. Res. 2002, 150, 694−705.
(12) Schrader, K. K.; Harries, M. D. A rapid bioassay for bactericides
against the catfish pathogens Edwardsiella ictaluri and Flavobacterium
columnare. Aquacult. Res. 2006, 37, 928−937.
(13) Ghosal, S.; Singh, S. K.; Srivastava, R. S. Chemical constituents
of Amaryllidaceae. Part 22. Alkaloids of Zephyranthes flava.
Phytochemistry 1986, 25, 1975−1978.
(14) Campbell, W. E.; Nair, J. J.; Gammon, D. W.; Codina, C.;
Bastida, J.; Villodomat, F.; Smith, P. J.; Albrecht, C. F. Bioactive
alkaloids from Brunsvigia radulosa. Phytochemistry 2000, 53, 587−591.
(15) Shoemaker, C. A.; Olivares-Fuster, O.; Arias, C. R.; Klesius, P.
H. Flavobacterium columnare genomovar influences mortality in
channel catfish (Ictalurus punctatus). Vet. Microbiol. 2008, 127, 353−
359.
(16) Evidente, A.; Cicala, M. R.; Randazzo, G.; Riccio, R.; Calabrese,
G.; Liso, R.; Arrigoni, O. Lycorine structure-activity relationships.
Phytochemistry 1983, 22, 2193−2196.
(17) Lamoral-Theys, D.; Andolfi, A.; Van Goietsenoven, D.;
Cimmino, A.; Le Calve,
́ ́
B.; Wauthoz, N.; Megalizzi, V.; Gras, T.;
Bruyere, C.; Dubois, J.; Mathieu, V.; Kornienko, A.; Kiss, R.; Evidente,
́
A. Lycorine, the main phenanthridine Amaryllidaceae alkaloid, exhibits
significant antitumor activity in cancer cells that display resistance to
proapoptotic stimuli: an investigation of structure-activity relationship
and mechanistic insight. J. Med. Chem. 2009, 52, 6244−6256.
(18) Schrader, K. K.; Andolfi, A.; Cantrell, C. L.; Cimmino, A.; Duke,
S. O.; Osbrink, W.; Wedge, D. E.; Evidente, A. A survey of phytotoxic
1183
dx.doi.org/10.1021/jf304586j | J. Agric. Food Chem. 2013, 61, 1179−1183