Am. J. Trop. Med. Hyg., 71(1), 2004, pp. 24-28
Copyright © 2004 by The American Society of Tropical Medicine and Hygiene
RELATIONSHIP OF ANTI-MICROBIAL ACTIVITY OF TETRACYCLINES TO THEIR ABILITY TO BLOCK THE L3 TO L4 MOLT OF THE HUMAN FILARIAL PARASITE BRUGIA MALAYI
T. V. RAJAN
Department of Pathology, University of Connecticut Health Center, Farmington, Connecticut
 |
ABSTRACT
|
The nematode parasites Wuchereria bancrofti, Brugia malayi, and B. timori cause a human disease known as lymphatic filariasis, which afflicts approximately 120 million people worldwide. These organisms are known to contain endosymbiotic bacteria (Wolbachia) that are related to rickettsiae. It has been previously reported that tetracycline blocks the L3 to L4 molt of the filarial parasite B. malayi, and suggested that this was related to their known anti-rickettsial activity. However, this interpretation was tempered by several observations. First, Wolbachia DNA could still be detected in nematodes from tetracycline-treated cultures. In addition, chloramphenicol, which has anti-rickettsial and anti-chlamydial activity, failed to inhibit the molt. These observations could not rule out the possibility that the anti-molting activity of tetracycline is due to pharmacologic activities unrelated to its anti-rickettsial functions. This study shows that chemically modified tetracycline, which does not to have anti-microbial activity, also blocks molting.
|
INTRODUCTION
|
Electron microscopic studies of several species of filarial nematodes have shown the presence of intracellular rickettsia-like organisms that are related to a group of arthropod endosymbionts known as Wolbachia.1 In filarial worms, these bacteria are concentrated primarily in the lateral cords of the hypodermis in all developmental stages, as well as in the ovary of adult females.2–4 They are transmitted vertically from one generation of nematodes to the next. Over the course of evolution, this mode of transmission is generally regarded as selecting for mutualistic interactions, since the survival of the microbe is entirely dependent on the ability of the host to survive and reproduce.5 Therefore, it is not unreasonable to hypothesize that the biology of Wolbachia and their filarial nematode hosts may be tightly intertwined. However, the precise nature of the interaction between Wolbachia and filarial worms has not been elucidated.
It has been observed that Mongolian jirds (Meriones unguiculatus) that are treated with tetracycline and simultaneously infected with Brugia pahangi have significantly decreased adult worm recoveries.6 The antibiotic had been initially used to treat an unrelated bacterial outbreak within the jird colony. Subsequent investigations showed that maximal reductions in adult worm burdens were achieved if the drug was given during both the L3 to L4 and the L4 to L5 molts.6 Since Wolbachia are related to rickettsiae and since tetracycline is known to be anti-rickettsial, it was proposed that elimination of the endosymbiont might be responsible for its observed effect on worm burdens.
Since maximal adult worm reductions were observed when jirds were exposed to tetracycline during the time the parasite would be undergoing its L3 to L4 molt, and since Wolbachia are present in the tissue responsible for synthesizing cuticle (the hypodermal syncitium), it was reasoned that the endosymbiont might have a critical role in the molting process of the filarial host. To investigate this theory, L3 larvae were treated in an in vitro system with a panel of antibiotics.7 Smith and Rajan found that tetracycline significantly inhibited the L3 to L4 molt. However, chloramphenicol, which has proven efficacy in human rickettsial and chlamydial infections, did not inhibit molting at concentrations equal to or exceeding the minimum inhibitory concentration of susceptible organisms. Furthermore, it was observed that Wolbachia DNA was still present in larvae that had been treated with tetracycline at concentrations that completely blocked molting.7 Therefore, although it was tentatively concluded that tetracycline might inhibit molting through its anti-microbial activity, Smith and Rajan reported that "the possibility remains that the observed molting inhibition is caused by other pharmacological actions of tetracycline."7 The family of compounds characterized by the four-ring tetracycline structure has diverse activities such as inhibition of matrix metalloprotease (MMP), prevention of pathogenic tissue destruction, and inhibition of cyclooxygenase-2 and nitric oxide synthase. Therefore, this cautious interpretation was not unreasonable.
Over the past few years, Greenwald and Golub have synthesized a number of derivatives of tetracyclines that lack the antimicrobial activity of the better known tetracycline family members.8 They found that removal of the dimethylamino group attached to the fourth carbon of the A ring of the tetracycline molecule abolishes antimicrobial activity. This compound, named CMT-1, still possessed the ability to inhibit collagenase. They have now synthesized a variety of compounds with increasing potency and bioavailability. These compounds have been recently renamed the Colseries.8
Since experiments with tetracycline and other antibiotics led to a less than definitive conclusion regarding the mechanism of inhibition of the molt, it was reasoned that the chemically modified tetracyclines (CMTs) could help resolve the issue. If they too block molting, one could reasonably conclude that the antimicrobial activity of tetracycline is not required to block the L3–L4 molt. Conversely, if the CMTs did not block the molt, one could reasonably conclude that the antimicrobial activity of the tetracyclines is required to block molting. I report here the results of using two of these CMTs, Col-3 and Col-8.
|
MATERIALS AND METHODS
|
Brugia larvae.
Most of the experiments reported here have been conducted with both B. malayi and B. pahangi L3. Larvae were obtained from the University of Georgia (Athens, GA) through a contract with the National Institute of Allergy and Infectious Diseases (Bethesda, MD). They were shipped in tissue culture medium supplemented with penicillin, streptomycin, and gentamicin.9 Upon arrival, the larvae were washed through several changes of clean, sterile tissue culture medium, and plated in 96-well tissue culture dishes at a concentration of five larvae per well.
Culture conditions.
Over the past few years, increasingly stringent conditions for obtaining the L3–L4 molt in vitro have been developed. Molting efficiencies range from 80% to 100%, depending upon the batch of larvae.10 Five larvae were cultured in each well of a 96-well cluster dish in a final volume of 200 µL of minimal essential medium (catalog #12571; In-vitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum. The plates were incubated at 37°C in a water-jacketed incubator maintained in an atmosphere of 5% CO2 and 95% air. Cultures were left undisturbed except for enumeration on day 2 of any larvae that might have died in transit or early in culture. These numbers were subtracted from the input of five larvae in the well to ensure that larvae dying early in culture do not skew the interpretation of the molting assay. On day five of initiation of culture, cultures were supplemented with ascorbic acid at a final concentration of 75 µM for B. malayi and 50 µM for B. pahangi. The plates were removed from the incubator on day 8, and visually inspected for clean cast cuticles. The number of cuticles divided by the number of fully viable L3s on day 2 per well is the percent molting observed per well. For each condition to be examined, a minimum of five or more wells was set up. The percent inhibition under any condition is expressed as the mean inhibition in this set of 5 or 10 wells in comparison to the mean percent molting in the set of 5–10 wells without the inhibiting agent.
Reagents.
Tetracycline hydrochloride and doxycycline were obtained from the Sigma Chemical Company (St. Louis, MO). Stock solutions at a concentration of 1 mg/mL were made in sterile distilled water and sterilized by filter sterilization through 0.2-µ pore size Miltex filters (Millipore, Billerica, MA). Col-3 and Col-8 were generous gifts from The Collagenex Corporation (Newtown, PA). Since these compounds are not soluble in aqueous solvents, they were dissolved in dimethylsulfoxide (DMSO) at a concentration of 5 mg/mL. They were diluted in sterile RPMI 1640 medium before use. The structures of the compounds used in this study are shown in Figure 1
.
View larger version (20K):
[in this window]
[in a new window]
|
FIGURE 1. Structures of the three tetracycline derivatives used in this experiment. Tetracycline is the parent compound. The four aromatic rings are designated A through D, with ring A at the far right. Col-3 and Col-8 were derived from tetracycline by removal of the dimethylamino group at position 4 of ring A.
|
|
Isolation of genomic DNA.
At the end of the culture (on day 8 or thereafter), all the viable L4 larvae were collected and washed with sterile phosphate-buffered saline to remove any serum components. They were then resuspended in a small volume of Tris-buffered saline. Sodium dodecyl sulfate, 2-mercaptoethanol, and protease K were added to final concentrations of 1%, 5%, and 10 µg/mL, respectively. The larvae were incubated overnight at 65°C and the genomic DNA was then precipitated with isopropanol. The DNA was then washed with 70% ethanol wash and dissolved in sterile Tris-EDTA buffer. DNA was also prepared from adult B. pahangi and B. malayi.
Amplification by a polymerase chain reaction (PCR).
To quantitate relative amounts of Wolbachia under various treatment conditions, semi-quantitative, duplex DNA amplification reactions were used with primers for a single copy B. malayi gene (BmCol111) and Wolbachia 18S ribosomal DNA (rDNA). The primers used for the B. malayi gene were 5'-GGGGAATTCGGATCCCTGGGTAGA-3' (sense strand) and 5'-CCCGAATTCGGATCCCTTTGGTCCTGG-3' (anti-sense strand). The primers for the Wolbachia ribosomal genes were 5'-TGAGGAAGATAATGACGG-3' (sense strand) and 5'-CCTCTATCCTCTTTCAACC-3' (anti-sense strand). Amplification was performed in a volume of 20 µL containing 1x PCR buffer (Invitrogen), 1.5 mM MgCl2, 200 µm dNTPs, 20 ng/mL of forward and reverse primers, and 2.5 units of recombinant Taq polymerase (Invitrogen). Samples were incubated for 25 cycles at 55°C for one minute, 72°C for one minute, and 94°C for one minute. At the end of 25 cycles, samples were incubated at 72°C for 15 minutes. The products were visualized after electrophoresis of 5 µL of the reaction mixture on 6% polyacrylamide gels stained with ethidium bromide.
|
RESULTS
|
Inhibition of the molting of B. malayi larvae by tetracy-cline.
As previously demonstrated,7 tetracycline consistently blocked the L3 to L4 molt of both B. malayi and B. pahangi. A typical dose-response curve, with tetracycline added to the culture on day 1, is shown in Figure 2. It is clearly seen that at low doses (10 µg/mL), the molting efficiency is equivalent to that seen in the absence of drug. At higher doses (> 40–50 µg/mL), tetracycline completely blocks molting. It is important to point out that even at these high doses, the viability of the larvae is not affected. While they do not molt, they continue to exhibit their characteristic vigorous, constant motion. Thus, the drug does not kill the larvae; it blocks only this specific morphogenic step.
View larger version (9K):
[in this window]
[in a new window]
|
FIGURE 2. Dose-response curve with tetracycline added to the culture on day 1. Larvae (L3) of Brugia malayi were incubated in minimal essential medium supplemented with 10% fetal bovine serum. Bars show the mean percent molting with the standard error of the mean. In the absence of any added drug, molting was observed at a high efficiency. When tetracycline was added to the culture on the day of initiation, molting was not significantly affected at final drug concentrations of 10 or 20 µg/mL. At higher concentrations, molting was inhibited until, at 50 µg/mL, it was completely inhibited. However, at this concentration, the worms continued to stay actively motile.
|
|
Blocking of molting by Col-3 and Col-8.
A dose-response curve of the effect of Col-3 on molting is shown in Figure 3. In initial experiments, Col-3 and Col-8 were used at doses comparable to that of tetracycline. At these high doses, both Col-3 and Col-8 rapidly killed the larvae; all were dead within the first 24 hours. Therefore, the doses of these two agents were titrated downward until a dose was obtained that inhibited molting without killing the larvae. As can be seen in Figure 3, Col-3 blocks molting at doses greater than 0.3 µg/mL without interfering with the viability of the larvae. Col-8 completely inhibits molting at a concentration of 1 µg/mL.
View larger version (10K):
[in this window]
[in a new window]
|
FIGURE 3. Dose-response curve of the effect of Col-3 on molting. Larvae (L3) of Brugia malayi were incubated in minimal essential medium supplemented with 10% fetal bovine serum. Bars show the mean percent molting with the standard error of the mean. In the absence of any added drug, molting was observed at a high efficiency. When Col-3 was added to the culture on the day of initiation, molting was not significantly affected at a final drug concentration of 0.1 µg/mL. At 0.3 µg/mL, it was completely inhibited. The worms continue to stay actively motile even at a concentration of 1 µg/mL.
|
|
Since DMSO was used as the solvent to test the effect of drugs on molting, one must repeatedly ensure that the solvent alone does not inhibit molting. In one representative experiment, the mean ± SD percent molting efficiencies in regular medium alone or medium supplemented with 0.1%, 0.2%, 0.5%, or 1% DMSO were 56 ± 16%, 66 ± 16%, 50 ± 30%, 56 ± 16%, and 72 ± 22%, respectively. These values are not statistically significantly different from each other. Thus, at levels (0.01–0.1%) that are present in the Col-3 and Col-8 solutions used, the solvent alone should have no deleterious effect on molting.
Presence of Wolbachia DNA in larvae treated with Col-3 and Col-8.
Since the amounts of DNA obtained from the L4 were too small to quantitate accurately, adult B. malayi and B. pahangi were treated under conditions that blocked the L3 to L4 molt. At the end of the culture conditions (day 8), DNA was isolated from the worms. The DNA samples were amplified with BmCol1 primers alone and the composition of the template was adjusted to ensure that the intensities of the bands were equalized. The PCR amplifications reactions were then performed with BmCol1 and Wolbachia primers. Representative data are shown in Figure 4. Under conditions where the bands for CW3 are roughly equivalent, the amount of Wolbachia DNA seen in the Col-3-treated culture is approximately the same as that in the control cultures.
View larger version (20K):
[in this window]
[in a new window]
|
FIGURE 4. Wolbachia DNA in larvae treated with Col-3 and Col-8 detected by a polymerase chain reaction. Brugia pahangi adult worms were incubated in minimal essential medium supplemented with 10% fetal bovine serum. The DNA was extracted from the worms as indicated in the Materials and Methods and amplified either with primers for Wolbachia ribosomal DNA (rDNA) (lanes 1-4) or B. malayi collagen 1 (lanes 5-8). The adult worms were grown in growth medium alone (lanes 1 and 5) or treated with 0.5 µg/mL of Col-3 (lanes 2 and 6), 1 µg/mL of Col-8 (lanes 3 and 7) or 50 µg/mL of tetracycline (lanes 4 and 8). The intensity of the bands was determined using Photoshop 5.0 (Adobe Systems, San Jose, CA). The data below the electrophoretic pattern indicate luminosity of the bands and the ratio of the intensity of the rDNA fragment to the Col-1 fragment. In all three instances, the amounts of rDNA expressed as a fraction of the CW3 fragment are comparable. Lane M = molecular mass markers.
|
|
|
DISCUSSION
|
Bosshardt and others6 reported that jirds treated with tetracycline from the initiation of filarial infection contained fewer adult filarial worms than expected at the time of necropsy. The discovery of the Wolbachia endosymbiont within the filariae provided a possible explanation for this observation. One could hypothesize that tetracycline was killing the endosymbiont, which plays a critical but undefined role in molting. Since the original experiments of Bosshardt and others seem to indicate that tetracycline had to be given relatively early in infection to reduce adult worm burdens, it was reasoned that they were inhibiting an early step in the morphogenesis of the filariae, perhaps the L3 to L4 molt. Earlier experiments confirmed that tetracycline also blocked this molt.7
However, several observations made in the course of these experiments led to an interpretation of the results with some reservations. Chloramphenicol, which has well documented anti-rickettsial efficacy, did not block the molt even at doses that have been demonstrated to have anti-rickettsial activity. Another finding that resulted in some reservations was that even in cultures that were completely inhibited from molting by tetracycline, Wolbachia DNA could still be detected.
Because of these reservations, our group sought to test the effect of CMTs that did not have antimicrobial activity, but still had anti-MMP and other metabolic activities attributed to tetracycline.8 There is a considerable amount of data in the literature to suggest that the CMTs do not have antimicrobial activity.8 Nonetheless, to confirm that these agents do not reduce the Wolbachia load in the filarial worms, the impact of these drugs on the amount of Wolbachia rDNA in the filarial worm during the treatment was determined. Recent data suggest that the number of Wolbachia within infective-stage larvae (L3) is small, and that there is a rapid increase in endosymbiont numbers in vivo, such that the L4 has abundant organisms within the hypodermis. Preliminary data indicate that this burst of Wolbachia replication does not take place under the culture conditions used (Rajan TV, unpublished data). As a result, it has been difficult to demonstrate the presence of Wolbachia DNA under these culture conditions. Therefore, B. pahangi and B. malayi adult worms were treated with the drugs under the same conditions used in the molting analyses The data presented in Figure 4
demonstrate that the Wolbachia rDNA content within the worms is not altered by such treatment. Thus, consistent with the literature reported over the last decade by Greenwald Golub,8 it appears that these drugs do not have Wolbachistatic activity under the conditions used in this study.
It important to note that treatment even with a known Wolbachistatic agent such as tetracycline does not eliminate Wolbachia DNA (Figure 4, lanes 4 and 8). Numerous experiments similar to those in Figure 4 have consistently shown that when tetracycline is tested as in this study (a single addition at an early time point in the culture), there is an inconsistent change in DNA content at the end of the culture. In some instances,7 there is some reduction in Wolbachia rDNA. More often, however, there is essentially no change in the content of Wolbachia rDNA in these cultures. In the experiments shown in Figure 4, treatment with tetracycline resulted in no change in the content of Wolbachia rDNA. At this point, however, it is important to point out that tetracycline was used under conditions different from those used by Hoerauf and others.12 They used doxycycline for six weeks. It is conceivable that the very short exposure to tetracycline might be responsible for the lack of elimination of Wolbachia in the cultures reported in this study.
In conclusion, the data in this report indicate that it is possible for a chemical with the basic four-ring structure of the tetracyclines to have anti-molting activity while being devoid of antimicrobial activity. The data do not rule out the possibility that the traditional antimicrobial tetracyclines function at least in part due to their ability to kill or inhibit Wolbachia. The efficacy of the CMTs, even in the absence of this antimicrobial activity, could be due to their more potent anti-MMP activity. It is also worth noting that the CMTs are more potent than tetracycline in blocking molting.
Received December 30, 2003.
Accepted for publication March 3, 2004.
Acknowledgments: I thank Carol McGuiness for excellent technical assistance and Jennifer Wegh for typing the manuscript. The idea of testing whether the CMTs could block molting was first suggested by Drs. Akhil and Ashok Vaidya.
Financial support: This work was supported by grants from the National Institutes of Health (AI-042362, AI-050228, and AI-039075).
Author’s address: T. V. Rajan, Department of Pathology, Room L-1037, University of Connecticut Health Center, 263 Farmington Avenue, Farmington CT 06030-3105, Telephone: 860-679-3221, Fax: 860-679-2936, E-mail: rajan{at}neuron.uchc.edu.
|
REFERENCES
|
- Bandi C, Anderson TJ, Genchi C, Blaxter ML, 1998. Phylogeny of Wolbachia filarial nematodes. Proc R Soc Lond B Biol Sci 265: 2407–2413.[Medline]
- Kozek WJ, 1977. Transovarially-transmitted intracellular micro-organisms in adult and larval stages of Brugia malayi. J Parasitol 63: 992–1000.[Medline]
- Kozek WJ, Marroquin HF, 1977. Intracytoplasmic bacteria in Onchocerca volvulus. Am J Trop Med Hyg 26: 663–678.
- McLaren DJ, Worms MJ, Laurence BR, Simpson MG, 1975. Micro-organisms in filarial larvae (Nematoda). Trans R Soc Trop Med Hyg 69: 509–514.[ISI][Medline]
- Werren JH, 1997. Wolbachia run amok. Proc Natl Acad Sci USA 94: 11154–11159.[Free Full Text]
- Bosshardt SC, McCall JW, Coleman SU, Jones KL, Petit TA, Klei TR, 1993. Prophylactic activity of tetracycline against Brugia pahangi infection in jirds (Meriones unguiculatus). J Parasitol 79: 775–777.[Medline]
- Smith HL, Rajan TV, 2000. Tetracycline inhibits development of the infective-stage larvae of filarial nematodes in vitro. Exp Parasitol 95: 265–270.[ISI][Medline]
- Greenwald R, Golub L, 2001. Biologic properties of non-antibiotic, chemically modified tetracyclines (CMTs): a structured, annotated bibliography. Curr Med Chem 8: 237–242.[ISI][Medline]
- Yates JA, Schmitz KA, Nelson FK, Rajan TV, 1994. Infectivity and normal development of third stage Brugia malayi maintained in vitro. J Parasitol 80: 891–894.[Medline]
- Rajan TV, Paciorkowski N, Kalajzic I, McGuiness C, 2003. Ascorbic acid is a requirement for the morphogenesis of the human filarial parasite, Brugia malayi. J. Parasitol 89: 868–870.[Medline]
- Caulagi VR, Rajan TV, 1995. The structural organization of an alpha 2 (type IV) basement membrane collagen gene from the filarial nematode Brugia malayi. Mol Biochem Parasitol 70: 227–229.[ISI][Medline]
- Hoerauf A, Adjei O, Buttner DW, 2002. Antibiotics for the treatment of onchocerciasis and other filarial infections. Curr Opin Investig Drugs 3: 533–537.[Medline]
TOP
ABSTRACT
INTRODUCTION
