In Vivo Antimalarial Activity of the Beta-Carboline Alkaloid Manzamine A

Abstract

Manzamine A, a β-carboline alkaloid present in several marine sponge species, inhibits the growth of the rodent malaria parasite Plasmodium berghei in vivo. More than 90% of the asexual erythrocytic stages of P. berghei were inhibited after a single intraperitoneal injection of manzamine A into infected mice. A remarkable aspect of manzamine A treatment is its ability to prolong the survival of highly parasitemic mice, with 40% recovery 60 days after a single injection. Oral administration of an oil suspension of manzamine A also produced significant reductions in parasitemia. The plasma manzamine A concentration peaked 4 h after injection and remained high even at 48 h. Morphological changes of P. berghei were observed 1 h after treatment of infected mice. (−)-8-Hydroxymanzamine A also displayed antimalarial activity, whereas manzamine F, a ketone analog of manzamine A, did not. Our results suggest that manzamine A and (−)-8-hydroxymanzamine A are promising new antimalarial agents.

Malaria remains the most devastating infectious parasitic disease, inflicting both death and economic losses on at least half the world's population. Numerous attempts have been made to control the disease by using vector control measures and/or chemoprophylaxis, but they have had limited success (18). Immunoprophylaxis holds promise, but effective vaccines are still not available. Presently, the most effective way of dealing with malaria is the administration of chemotherapeutic agents. Although drug treatments of malaria are currently the best means of disease management, there is an urgent need for the development of structurally novel and effective antimalarial drugs because of increasing resistance to most presently available antimalarial drugs (15, 16, 19).

Some of the most effective antimalarial drugs available, quinine and artemisinin, are natural products derived from terrestrial plants. However, recent research suggests that marine organisms may also produce compounds with activity against malaria parasites (4, 11, 21). Manzamines are a structurally unique group of β-carboline alkaloids isolated from several marine sponge species found in waters of the Indian Ocean and the Pacific Ocean. Manzamine A (Fig. 1) was initially isolated from a Haliclona sp. (17) but has been subsequently found in other genera of marine sponges, including Pellina (14), Pachypellina (7), Xestospongia (3, 8), Ircinia (10), and Amphimedon (9). In addition, more than 30 other compounds structurally related to manzamine A have been isolated from sponges and characterized; these include 8-hydroxymanzamine A and the ketone derivative manzamine F (Fig. 1). The origin, isolation, and chemistry of various manzamines have been reviewed (6, 13), with the complete synthesis of manzamine A being recently reported (20). The manzamines previously received considerable interest because of their potential as anticancer agents, with both manzamine A and manzamine F inhibiting the growth of P-388 mouse leukemia cells (6) and 8-hydroxymanzamine A showing moderate cytotoxicity against KB and LoVo cells (7). We have now found that manzamine A and 8-hydroxymanzamine A significantly inhibit malaria parasites not only in vitro but also in vivo.

FIG. 1.

Structures of manzamines tested for antimalarial activity. The structures of manzamine A (A), (−)-8-hydroxymanzamine A (B), and manzamine F (C) are shown.

MATERIALS AND METHODS

Compounds.

The test compounds manzamine A (isolated as a monohydrochloride salt) and manzamine F were provided by T. Higa, University of the Ryukyus. (−)-8-Hydroxymanzamine A, the enantiomer of 8-hydroxymanzamine A, was provided by M. T. Hamann, University of Mississippi. The reference drug artemisinin was provided by R. Haynes, and the reference drug chloroquine (available as a diphosphate salt) was purchased from Sigma Chemical Co., St. Louis, Mo.

Animals and parasites.

Four-week-old, male Swiss albino mice were purchased from the Laboratory Animal Centre, Singapore, Singapore. Mice were kept in cages in an animal room with half-day alternating light and dark periods. Tap water and mouse feed were provided ad libitum. Plasmodium berghei (ANKA strain) erythrocytic stages were maintained by serial passaging of infected blood in male Swiss albino mice.

Infection and treatment.

Four-week-old, male Swiss albino mice were injected intraperitoneally with 10⁷ P. berghei-infected mouse erythrocytes. On day 2 after infection, mice were treated with a single intraperitoneal injection of either the test compound or a reference drug (chloroquine or artemisinin) within a concentration range of 50 to 1,000 μmol/kg of body weight. All test compounds and reference drugs were injected as a suspension in 5% Tween 60 saline. For oral administration, the test compounds were given as a suspension in corn oil, and mice were given two consecutive doses of test compound at 100 μmol/kg on days 2 and 3 after infection. Control mice received only 5% Tween 60 saline or corn oil. Survival of mice was recorded daily. Percent parasitemia was determined microscopically (magnification, ×1,000) from mouse tail blood smears that were fixed with methanol and stained with Giemsa stain.

Transmission electron microscopy.

Blood samples from individual P. berghei-infected mice were collected at various times after a single intraperitoneal treatment of 100 μmol of manzamine A per kg on day 3 after infection. For transmission electron microscopy, approximately 0.5-ml blood samples were fixed with 3% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4), postfixed with 1% osmium tetroxide and then 1% uranyl acetate, dehydrated in a graded ethanol series, and embedded in Spurr's resin. The resulting blocks were cut by using a Reichert-Jung Ultracut with a glass knife. Ultrathin sections were mounted on 150-mesh copper grids, stained with 1% uranyl acetate and 1% lead citrate, and examined with a JEM-100CX electron microscope.

Detection of manzamine A in mouse plasma.

Manzamine A in plasma was detected by liquid chromatography-selected reaction monitoring-mass spectrometry (LC-SRM-MS). P. berghei-infected mice were treated with manzamine A (100 μmol/kg intraperitoneally) on day 3 postinfection. Blood samples were collected from individual mice at specific times up to 48 h posttreatment. Plasma was isolated by centrifugation and extracted with 95% acetonitrile containing 5 mM ammonium acetate. The plasma extract was filtered, and 5-μl samples were injected into a Shimadzu LC-10 AD microbore high-pressure liquid chromatograph interfaced with a Perkin-Elmer API 300 turbo-ion-spray tandem mass spectrometer. Samples were separated on a Prodigy octyldecyl silane column (30 by 1 mm, 5-μm inner diameter) by elution at 50 μl/min with 86% acetonitrile containing 5 mM ammonium acetate. Manzamine A was detected by monitoring the transition of the protonated manzamine A precursor ion from m/z 549.5 (M + H)⁺ to m/z 531.5 (M + H − H₂O)⁺. Peak areas for the product ion chromatograms were integrated, and concentrations were determined from a linear calibration curve for manzamine A in spiked plasma. The limit of detection for manzamine A in plasma by LC-SRM-MS was 2.5 pg.

Statistical analysis.

Results were subjected to a nonparametric statistical assessment with a confidence interval of 99% (P < 0.01) by using Analyse-it software linked to Microsoft Excel. The program uses the Mann-Whitney U test to determine significant differences between two groups of data.

RESULTS

The survival over time (60 days posttreatment) of mice infected with the erythrocytic stages of P. berghei was compared to survival after treatment with a single intraperitoneal injection of either manzamine A, (−)-8-hydroxymanzamine A, manzamine F, chloroquine, or artemisinin (Table 1). All control mice and mice treated with manzamine F died within 4 days posttreatment. In contrast, a single intraperitoneal administration of manzamine A (50 or 100 μmol/kg) or (−)-8-hydroxymanzamine A (100 μmol/kg) prolonged the survival of P. berghei-infected mice for more than 10 days, with 40% of mice treated with 100 μmol of manzamine A per kg surviving for more than 60 days and recovering with no detectable parasitemia. Similarly, oral administration of an oil suspension of manzamine A or (−)-8-hydroxymanzamine A significantly prolonged the survival of infected mice. The ability of manzamine A and its hydroxyl derivative to extend the lives of infected mice far exceeded that of chloroquine and artemisinin, two of the most important human therapeutic antimalarial drugs. Manzamine A was toxic to mice at 500 μmol/kg, but it showed more slowly acting toxicity than chloroquine, which caused almost instantaneous death of the treated mice at 500 μmol/kg.

TABLE 1.

Survival over time (60 days posttreatment) of P. berghei-infected mice treated intraperitoneally (i.p.) or orally with test compounds or reference drugs

Treatment	Route	Dose (μmol/kg)	No. of mice tested	No. of mice surviving on the following day after treatment:
				0	2	4	6	10	15	25	60
Manzamine A	i.p.	1,000a	4	4	1	0
	i.p.	500a	4	4	2	0
	i.p.	100b	5	5	5	5	5	5	4	2	2
	i.p.	50b	5	5	5	5	5	5	3	0
	Oral	100bc	5	5	5	5	5	4	0
(−)-8-Hydroxymanzamine A	i.p.	100b	5	5	5	5	5	5	0
	Oral	100bc	4	4	4	4	4	4	0
Manzamine F	i.p.	100	5	5	5	0
Artemisinin	i.p.	1,000b	5	5	5	5	5	5	1	1	1
	i.p.	500b	5	5	5	5	5	1	0
	i.p.	100b	5	5	5	4	0
	i.p.	50	5	5	5	2	0
Chloroquine	i.p.	1,000a	2	0
	i.p.	500a	2	0
	i.p.	100b	5	5	5	5	5	0
	i.p.	50b	5	5	5	5	5	0
Control			14	14	14	0

^aMice died due to toxicity upon treatment. 

^bSignificant difference (longer survival time) between this treatment and the control (P < 0.01). 

^cTwo doses (100 μmol/kg each). 

Table 2 shows the percent parasitemia in mice receiving different treatments for the first 3 days after the treatments. The mean percent parasitemia of the control mice increased drastically within 3 days after treatment, from 2.1% on day 1 to 83.5% on day 3, with fatality by day 4 after treatment. A single intraperitoneal injection of manzamine A (50 or 100 μmol/kg) or (−)-8-hydroxymanzamine A (100 μmol/kg) reduced the parasitemia in mice by more than 90% compared to that in control mice for the first 3 days after treatment. Such suppressive activity is comparable to that of chloroquine and superior to that of artemisinin at the same dose. However, intraperitoneal treatment with manzamine F failed to show any inhibitory effect, as the rise of parasitemia in manzamine F-treated mice showed no significant difference from that in control mice. Oral administration (two doses of 100 μmol/kg) of manzamine A or (−)-8-hydroxymanzamine A also produced more than 90% inhibition of parasitemia compared to that in control mice for the first 3 days after the first treatment.

TABLE 2.

Percent parasitemia in mice receiving different treatments for the first 3 days after the treatments

Treatment	Routea	Dose (μmol/kg)	No. of mice tested	% Parasitemia (mean ± SD) on the following day after treatment:
				0	1	2	3
Control			14	2.1 ± 1.0	22.9 ± 8.6	57.7 ± 9.9	83.5 ± 8.2
Manzamine A	i.p.	100	5	1.4 ± 0.6	1.0 ± 1.3c	1.8 ± 2.8c	3.1 ± 4.3c
	i.p.	50	5	1.8 ± 0.4	1.7 ± 1.3c	2.1 ± 1.7c	3.4 ± 2.3c
	Oral	100b	5	1.2 ± 0.4	2.7 ± 1.5c	3.2 ± 1.9c	5.5 ± 6.6c
(−)-8-Hydroxymanzamine A	i.p.	100	5	1.7 ± 0.7	0.2 ± 0.2c	0.2 ± 0.2c	2.1 ± 4.7c
	Oral	100b	4	2.5 ± 0.8	4.4 ± 1.6c	2.2 ± 1.5c	7.1 ± 6.0c
Manzamine F	i.p.	100	5	3.6 ± 0.6	22.8 ± 5.1	63.1 ± 14.9	82.6 ± 7.8
Chloroquine	i.p.	100	5	2.1 ± 0.8	1.1 ± 0.4c	0.2 ± 0.1c	0.1 ± 0.03c
	i.p.	50	5	1.7 ± 0.3	1.0 ± 0.6c	0.3 ± 0.2c	0.1 ± 0.03c
Artemisinin	i.p.	100	5	1.2 ± 0.5	2.3 ± 1.8c	10.4 ± 11.7c	35.9 ± 20.7c
	i.p.	50	5	1.4 ± 0.2	4.7 ± 4.8c	22.2 ± 20.1c	47.0 ± 23.3c

^ai.p., intraperitoneal. 

^bSee Table 1, footnote c. 

^cSignificant difference between this treatment and the control (P < 0.01). 

Transmission electron microscopy revealed progressive changes in the morphology of the erythrocytic forms of P. berghei parasites after intraperitoneal administration of manzamine A (Fig. 2A to D), with initial changes seen only 1 h after treatment. One hour of exposure to manzamine A induced the formation of membrane-bound vesicles with various electron densities within some parasites (Fig. 2B). Four hours after treatment, other morphological changes, such as the development of electron-dense vesicles, were observed in more parasites (Fig. 2C). By 12 h after drug exposure, the parasite cytoplasm showed marked degeneration and was filled with electron-dense vesicles (Fig. 2D). Almost all parasites had degenerated by 24 h after exposure to manzamine A. These morphological changes of P. berghei after treatment with manzamine A resemble those reported for chloroquine treatment (12).

FIG. 2.

Transmission electron micrographs of P. berghei treated with manzamine A. (A) Mouse erythrocyte containing untreated P. berghei. Magnification, ×19,000. (B) P. berghei 1 h after administration of manzamine A. Magnification, ×29,000. Electron-dense vesicles are present within the cytoplasm of the parasite (arrows). (C) P. berghei 4 h after manzamine A administration. Magnification, ×29,000. (D) P. berghei 12 h after treatment. Magnification, ×19,000. Increasing numbers of vesicles are present within the parasite.

Following intraperitoneal administration, manzamine A was detected and its levels rose sharply in the plasma of P. berghei-infected mice within 2 h after treatment, reaching a peak concentration in plasma at 4 h (Fig. 3). The level of manzamine A in mouse plasma fell slightly 4 h after administration but remained consistent for the next 44 h, with approximately 50% of the maximum concentration in plasma still being present in manzamine A-treated mice 48 h after administration.

FIG. 3.

Plasma manzamine A concentration (mean ± standard error) following intraperitoneal administration of 100 μmol/kg into P. berghei-infected mice (n = 5 for each time point).

DISCUSSION

The P. berghei-infected mouse model has been widely used as a preliminary test for the in vivo activity of potential antimalarial agents, as it provides a preclinical indication of any in vivo potential bioactivity as well as possible toxicity of the sample tested (1, 2, 5). Results from this study strongly indicate that manzamine A and its hydroxyl derivative, (−)-8-hydroxymanzamine A, are active against the asexual erythrocytic stages of P. berghei. All mice treated with a single dose (50 or 100 μmol/kg) of manzamine A, (−)-8-hydroxymanzamine A, chloroquine, or artemisinin experienced a recurrence of parasites despite the initial suppression of parasitemia development. However, in contrast to chloroquine- and artemisinin-treated mice, most infected mice treated with manzamine A were able to survive for a longer period of time carrying fulminating recurrent parasitemia; two mice, treated with 100 μmol of manzamine A per kg, were able to recover and clear the parasitemia completely. We speculate that immune responses developed in the host animals after treatment with manzamine A. These potential immunostimulatory effects of manzamine A warrant extensive further investigations. In addition to in vivo studies with mice, the effectiveness of manzamine A has also been tested in vitro against chloroquine-resistant clone W2 and mefloquine-resistant clone D6 strains of the human malaria parasite P. falciparum, with 50% inhibitory concentrations of <528.8 ng/ml (M. T. Hamann, personal communication).

The lack of antimalarial activity of manzamine F in comparison to manzamine A and (−)-8-hydroxymanzamine A provides some interesting information on structure-activity relationships. In contrast to manzamine A, manzamine F possesses a ketonic carbonyl group at C-31 and a hydroxyl group at C-8 but lacks a double bond between C-32 and C-33 and a hydrogen at N-27 (Fig. 1). However, the hydroxyl group at C-8 is present in the antimalarial compound (−)-8-hydroxymanzamine A. Therefore, the structural difference between the antimalarial manzamines [manzamine A and (−)-8-hydroxymanzamine A] and nonantimalarial manzamine F is limited to the eight-member ring from N-27 to C-34. The availability of other manzamines for testing in the near future may help elucidate the active chemical entity responsible for the antimalarial effects observed.

The pharmacokinetic profile of manzamine A in mouse plasma is consistent with the onset of its action against P. berghei in vivo. The rapid bioavailability of manzamine A in mice, within 2 h after treatment, correlates well with the early morphological changes seen in P. berghei, at 1 h after treatment. The continuous sustained level of manzamine A in mouse plasma ensures effective deterioration of the parasites. Further challenges will be to determine if the parent compound, manzamine A, gives rise to other active metabolites once released into biological fluids.

We suggest that manzamine A and selected derivatives or precursors of manzamine A have potential as novel antimalarial agents with rapid onset of action and prolonged antiparasitic activity. However, manzamine A is lethal to mice at a dose (500 μmol/kg) which is only 10 times higher than the dose (50 μmol/kg) that suppresses parasitemia and prolongs survival but does not cure. Despite the narrow therapeutic index of manzamine A seen in the above in vivo tests, the recent complete synthesis of manzamine A (20) will enable extensive structure-activity evaluation and synthesis of more effective and safer manzamine-related antimalarial compounds.