| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Biomedical Sciences, Iowa State University, Ames, Iowa, USA
2Correspondence: Department of Biomedical Sciences, #2008, College of Veterinary Medicine, Iowa State University, Ames, Iowa 50011-1250. E-mail: rjmartin{at}iastate.edu
ABSTRACT
Pharmacological experiments on Ascaris suum have demonstrated the presence of three (N-, L-, and B-) subtypes of cholinergic receptor mediating contraction of body wall muscle in parasitic nematodes (1)
. In the present study, these ionotropic acetylcholine (ACh) receptors (nAChRs) were activated by levamisole and bephenium under patch-clamp conditions and competitively antagonized by paraherquamide and 2-desoxoparaherquamide. A number of recordings exhibited three separate current amplitude levels, indicating the presence of small, intermediate, and large conductance subtypes of receptor. The mean conductance of the small conductance subtype, G25, was 22 ± 1 pS; the intermediate conductance channel, G35, was 33 ± 1 pS; and the large conductance channel, G45, was 45 ± 1 pS. The small channel was not antagonized significantly by paraherquamide and was identified as the N-subtype. The intermediate channel was preferentially activated by levamisole rather than bephenium and antagonized by paraherquamide: the intermediate channel was identified as the L-subtype. The large conductance channel was preferentially activated by bephenium, antagonized more by 2-desoxoparaherquamde than by paraherquamide and was identified as the B-subtype. These observations reveal that the three channel subtypes have different selectivity for cholinergic anthelmintics. The different selectivity of these compounds should be considered when dealing with drug resistant infections.—Qian, H., Martin, R. J., and Robertson, A. P. Pharmacology of N-, L-, and B-subtypes of nematode nAChR resolved at the single-chain level in Ascaris Suum.
Key Words: anthelmintic • levamisole • nicotinic • ACh receptor • patch-clamp
OVER ONE-THIRD OF THE HUMAN population is infected with intestinal parasites. In a World Health Organization report (2)
, it was observed that 17.3 million deaths, of the global total of 52.2 million deaths, were due to infectious and parasitic diseases. A common nematode parasitic disease of humans is ascariasis, which has a prevalence of 30–60% in endemic areas, producing symptoms of malnutrition, retardation of growth in children, diarrhea, and abdominal pain and in a smaller proportion of cases, death (3)
. Gastrointestinal nematodes are the major cause of morbidity in schoolchildren when measured in disability adjusted life years (DALYs). Levamisole is a representative of an important group of anthelmintic drugs (including pyrantel) that is used for treatment of ascariasis as well as other human nematode parasite infections (4)
. Members of this group act rapidly by selectively gating pharmacologically distinctive ionotropic (nicotinic) ACh receptor ion-channels (nAChR) on the body muscles of nematodes (5
, 6)
.
Human studies show that anthelmintic treatment with cholinergic anthelmintics and other anthelmintics is <100% effective. Albonico et al. (7)
described the efficacy of cholinergic anthelmintics against Ascaris sp. (roundworm), Trichuris sp. (whipworm) and Ancylostoma sp. (hookworm). The cure rates for these parasite species is <100%. A proportion of parasites is innately resistant and unaffected by treatment. The regular use of anthelmintics produces a Darwinian selection pressure that increases the proportion of resistant individuals in the population (acquired resistance). This has been well established for nematode parasites of domestic animals. More recently, significant resistance of human parasites to cholinergic anthelmintics has been described making the drugs ineffective (8
, 9)
. The prospect for the development of new anthelmintics is very limited, so there is an urgent need to investigate the mechanisms of anthelmintic resistance and to find new ways to counter resistance.
We have focused on sites of action of the cholinergic anthelmintics to gain a better understanding of their modes of action and mechanisms of resistance. In contraction assays using muscle strips from the parasitic nematode Ascaris suum, we have observed that there are pharmacologically distinct nAChRs that can be activated by the different cholinergic anthelmintics (1)
. We were able to separate three types of nAChR: the N-subtype that is preferentially activated by nicotine; the L-subtype that is preferentially activated by levamisole and antagonized by paraherquamide; and the B-subtype that is preferentially activated by bephenium and antagonized by paraherquamide and 2-desoxoparaherquamide. Our studies also found that levamisole resistance is associated with a loss of sensitivity of the L-subtype receptors with no loss in the sensitivity of the N-subtype of receptors (10
, 11)
. These observations suggest that N-type and B-type selective cholinergic agonists may be useful for overcoming some types of levamisole resistance.
In addition to the muscle contraction assays, we have also compared the effects of nicotine and levamisole under patch clamp (12)
to test our hypothesis that the selectivity of these two ligands is different. We found that both nicotine- and levamisole-activated channels had a wide and overlapping conductance range but that nicotine preferentially activates smaller, 26 pS, channels and that levamisole preferentially activates another, 39 pS, group of channels. In this study, we are able to separate out a third subtype to compare the effects of levamisole and bephenium at the single-channel level and test the effects of the novel competitive-antagonists paraherquamide and 2-desoxoparaherquamide (13
, 14
, 15)
. We were thereby able to identify the single-channel properties of the N-, L-, and B-receptors. These observations are notable because they emphasize the different selectivity of the different cholinergic anthelmintics between the different subtypes of cholinergic receptor. This means, for example, that loss of the L-subtype with levamisole resistance (15
, 11)
might be overcome by using other cholinergic agonists (methyridine: N-subtype) or antagonists (2-desoxoparaherquamide; B-subtype) with selectivity for other subtypes of nAChR.
MATERIALS AND METHODS
Maintenance of Ascaris suum
Adult Ascaris suum were collected from the IBP Meat Packing Plant (Storm Lake, IA). The worms were maintained at 32°C in Locke’s solution (changed daily) for no longer than 5 days. Locke’s solution contains (mM): NaCl, 155; KCl, 5; CaCl2, 2; NaHCO3, 1.5; D-glucose, 5. Mature, active Ascaris suum 
10–20 cm in length were selected for experiments.
Vesicle Preparation
The muscle membrane vesicle preparation has been described previously (5)
and is briefly outlined here. The Ascaris were dissected and a muscle flap was prepared and pinned cuticle side down onto a plastic dish lined with Sylgard. The muscle flap preparation was washed with maintenance solution to remove fragments of the gut. The maintenance solution was then replaced with collagenase solution. After collagenase treatment for 4–8 min at 37°C, the muscle preparation was washed and incubated at 37°C. Small membranous vesicles, 10–50 µm in diameter, grew out from the membrane of the muscle cells. These membranous vesicles were transferred to a recording chamber.
Electrophysiology
The maintenance solution in the chamber was replaced with bath solution containing (mM): CsCl, 35; Cs acetate, 105; MgCl2, 2; HEPES, 10; EGTA, 1; pH 7.2 with CsOH, at room temperature. The vesicle preparation was used within 3 h.
The patch-clamp technique was used to record the single-channel currents activated by levamisole or bephenium from the vesicle preparation. The pipettes were filled with pipette solution that contained (mM): CsCl, 140; MgCl2, 2; HEPES, 10; EGTA, 1; pH 7.2 with CsOH. Pipettes with resistances of 3–5 M
were used. The current signal was amplified by an Axopatch 200B amplifier (Axon Instruments, Union City, CA) filtered at 2 kHz (3-pole Bessel) and then sampled at 25 kHz digitized with a Digidata 1320A (Axon Instruments) and stored on a computer hard disk.
Data analysis
Data were idealized and analyzed using pCLAMP Ver 8.2 software (Axon Instruments). The generally low level of channel opening produced extremely rare multiple opening events; these were omitted from calculations of Popen and mean open time. The current amplitude histograms were fitted with the sum of one, two, and three Gaussian equation to determine the mean current amplitudes of the channel openings. The best Gaussian fits (confidence level>0.95) were used to help determine the number of amplitude populations present in one patch. This analysis and curve fitting were done using pCLAMP 8.2 and Prism V4 (GraphPad Software, San Diego, CA). To fit the distribution of the all the individual levamisole channel conductance, we used a simplex method to minimize the residual sum of squares with the aid of the NAG subroutine E040CCF to fit the sum of three Gaussian distributions (15)
and were able to obtain estimates of the areas, mean conductance, and SD of the three populations.
We measured Popen as the patch-Popen (the proportion of time channels in the patch were open during the recording). Results are mean ± SE and were tested for normality. Although the distribution of open times were exponential, we found that the distributions of the means of open times from different patches did not deviate significantly from normality so differences between three groups of means was tested by ANOVA and differences between two means were evaluated using t test.
Drugs
Paraherquamide and 2-desoxoparaherquamide were obtained from Pfizer Animal Health (Kalamazoo, MI). Other drugs and chemicals were purchased from Sigma Chemical Co (St. Louis, MO). Paraherquamide or 2-desoxoparaherquamide, dissolved in dimethyl sulfoxide (DMSO), was added to the bath solution to give final concentrations of 0.3–10 µM. The concentration of DMSO in bath was always <0.3%. Levamisole/bephenium were present and in the pipette solution to activate the channels.
RESULTS
Levamisole activated conductance of three different channels
When levamisole (30 µM) was used as the agonist in the pipette solution, single-channel currents (Fig. 1
A) with a conductance of 18–53 pS and mean open times of 0.2 – 2 ms were observed at membrane potentials of ±75 and ±50 mV from inside-out patches. These channels were never recorded in the absence of levamisole. To confirm that these single-channel currents were indeed activated by levamisole, a series of alternating patch recordings was made from the same "active" vesicle: first with, and then without, levamisole present in the pipette solution. In six such paired recordings, channels were recorded in the presence of levamisole but not in the absence of levamisole. On four occasions, a third patch recording was made from the same "active" vesicle with pipette solution again containing levamisole. Under these conditions, characteristic single-channel currents were observed in all four of four patches. These observations confirmed that the currents were activated by levamisole.
|
In each patch, levamisole activated up to three populations of channel currents that were clearly separated by differences in their conductance. Figure 1A
shows some representative channel openings at +75 mV in a single patch recording and the amplitude histogram that was fitted by the sum of three Gaussian distributions with mean currents of 2.0, 2.7, and 3.6 pA. In all experiments where three or more voltages were tested between +75 mV and –75mV, it was found that the current-voltage relationships of the channels were linear. In some experiments where only two potentials were available, it was also taken that the relationship was linear. We determined conductance from the slopes of the current-voltage plots using linear regression. In Fig. 1B
, the channels had a conductance of 26, 36, and 48 pS.
We recorded levamisole-activated currents from 58 such channels from 38 patches: 22 had only 1 conductance level present, 12 had 2 conductance levels present, and 4 had 3 conductance levels present. The conductance ranged from 18 to 53 pS. The wide range of conductance values coupled with the low error associated with the measurement of individual channel conductance (SE
3 pS, r2>0.95 with currents measured at 3 or more potentials) made it unlikely that there was only a single population of channels present. Since the experimental conditions were equal over an individual patch, the presence of the three types of conductance in single patches revealed the presence of separate and distinctive small-, intermediate-, and large-conductance channel populations. To determine the conductance ranges of the three populations, we determined the mean and 95% confidence intervals of the small, intermediate and large conductance populations from the 4 patches showing the three types of conductance (Table 1
). The confidence limits for the small, 22 pS, channels were 17.8 and 26.3 pS; the confidence limits for the intermediate, 32.7 pS, channels were 26.5 and 38.9pS; and the confidence limits for the large, 42.9 pS, channels were 36.6 and 49.3 pS.
|
Next, we classified the remaining channel conductance in the other single or double channel patches into three populations according to their conductance using the 95% confidence intervals determined in the three channel patches. Channels classified as small all had a conductance of <26.5 pS; channels classified as intermediate had a conductance of 26.5–36.6 pS; and channels classified as large had a conductance of >38.9 pS. Only two types of conductance that had values between 36.6 and 38.9 could not be classified and were excluded in the following analysis.
Table 1
shows that the mean ± SE of the conductance for all the small channels was 23 ± 1 pS (n=15), and for convenience we refer to this channel as G25. The mean ± SE of the conductance for all the intermediate channels was 33 ± 1 pS (n=22), and we refer to this channel population as G35. The mean ± SE of the conductance for all the large channels was 45 ± 1 pS (n=19), and we refer to this channel as G45.
Levamisole preferentially activates G35 channels
The confidence interval method showed that levamisole preferentially activated G35 channels over the other subtypes: 27% of the channels were G25 (15 of 56), 39% were G35 (22 of 56), and 34% were G45 (19 of 56). To estimate the proportion of each subtype present using another approach, we fitted the conductance histogram of all 58 channels activated by levamisole to the sum of three Gaussian distributions using the individual 58 conductance values and an iterative simplex method (58; Fig. 2
A); 28% of these channels were G25 channels with a mean ± SD of 24.5 ± 4.3 pS; 58% of the channels were G35 with a mean ± SD of 36.7 ± 5.3 pS; and 14% were G45 with a mean ± SD of 49.9 ± 1.6 pS. Both analytical approaches separated the channel subtypes and showed that levamisole activates more G35 channels than the other receptor channel subtypes. The average conductance of the G35 channels is close to channels identified as the L-subtype of receptor and the conductance of the G25 channels is the same as N-subtype of nAChRs (12)
.
|
Bephenium activates only G35 and G45 channels
When levamisole in the pipette solution was replaced by 1 or 10 µM bephenium (a cholinomimetic anthelmintic), only two channel types, G35 and G45, were observed in the patches. Bephenium did not activate the G25 channels. In 7 of the 19 patches with 1 µM bephenium in the pipette, G35 and G45 were recorded simultaneously; in the other patches (12 in 19 patches), only the G35 or G45 channel was present. The conductance range of all these channels was from 31 to 49 pS. Using the same approach that we used for the levamisole-activated channel currents, we again evaluated the mean conductance and 95% confidence intervals for the G35 and G45 channels. From the two channel recordings in the seven patches, we found that the G35 channel had a 95% confidence interval of 31.6–40.2 pS and the G45 channel had a 95% confidence interval of 41.6–47.1 pS. We used this information to sort the conductance of the single channel patches: conductance <40.2 pS was classified as G35 channels; channels greater than 41.6 pS was classified as G45 conductance channels. Overall we found that the G35 channels activated by bephenium had a conductance of 36 ± 1 pS (mean±SE; n=10). We found that the conductance of the G45 activated by bephenium had a conductance of 43 ± 1 pS (mean±SE; n=16). The means and the conductance ranges of these two bephenium-activated populations are similar to those of the G35 and G45 channels activated by levamisole.
Bephenium activates more G45 than G35 channels
The confidence interval method revealed that when 1 µM bephenium was the agonist 38% of the channel population was G35 (10 of 26) and 62% was G45 (16 of 26). When the agonist was 10 µM bephenium, the proportion of G35 channels was 42% (8 of 19) and the proportion of G45 channels was 58% (11 of 19). Thus bephenium activated more G45 channels than G35 channels, suggesting that these channels represent the B-subtype of receptors (1)
.
Different subtypes have different mean open times
Once we had separated the channels activated by 30 µM levamisole into G24 G35, and G45, we were able to determine their mean open times by fitting a single exponential distribution to events >0.5 ms. We set the patch potential to +75 mV for these observations to minimize open channel block. In multiple conductance patches, each channel opening was separated into G25, G35, and G45 and their mean open times determined separately. An example of the three open time histograms from one patch is shown in Fig. 3
A. The G25 population had the shortest mean open times, and the G45 population had the longest mean open times. The G25 population had a mean open time of 0.6 ± 0.1 ms (mean±SE; n=14); the mean open time of the G35 population was 0.8 ± 0.1 ms (mean±SE; n=18); and the mean open time of the G45 population was 1.2 ± 0.1 ms (mean±SE; n=17). The distributions of the mean open times of the three populations were not significantly different from a normal distribution. ANOVA was used to test the difference between the three (G25 G35, G45) groups and significance was found (F=8.9, df 2, P<0.005). A t test was used to test the differences between the means of the G25 and G35 groups (P<0.03), the means of the G35, and G45 (P<0.03), and the means of the G25 and G45 groups (P<0.001): they were all significant (Fig. 3C
).
|
A similar difference between the G35 and G45 channels was observed when bephenium was the agonist, Fig 3B
. The mean open times of the G35 channels activated by 1 µM bephenium was 1.1 ± 0.2 ms (mean±SE; n=7), and this was significantly (P<0.05) shorter than the mean open time of the G45 population that was 2.4 ± 0.2 ms (mean±SE; n=11) when tested with a t test, Fig. 3C
. The differences between the mean open times of the different conductance populations are consistent with these channel populations belonging to separable distinctive subtypes groups and not being a single continuum.
Bath application of paraherquamide inhibits the opening of nAChRs
Our earlier muscle contraction experiments have demonstrated that paraherquamide is a competitive antagonist (1)
. It is not possible to make outside-out patch recordings from the Ascaris muscle vesicles because they are under internal pressure and burst when whole-cell recordings are attempted. Fortunately, paraherquamide is lipid soluble and passes from the bath through inside-out patches to reach the extracellular receptor to act as an antagonist (Fig. 4
A). This is a technique that we have established and used for the application of lipid soluble ligands (16
, 12)
. Here we measured and normalized the antagonism using paired patches. We recorded from two patches and measured the Popen values using the same protocol and time course to control for any rundown or effects of adding solution to the inner membrane. In the first control inside-out patch with channels activated by 30 µM levamisole in the pipette, we tested the effect of adding antagonist-free solution to the bath; there was sometimes a small decrease in Popen in the control patch. In the second test patch, with channels also activated by 30 µM levamisole in the pipette, we observed the effect on channel opening of adding paraherquamide to the bath. As a normalized measure of inhibition, the percent change in Popen produced by paraherquamide was divided by the percent change in Popen produced by the antagonist-free solution.
|
The inhibitory effect of paraherquamide was dose dependent (Fig. 4B
) and was described by the logistic function with an IC50 of 3.1 µM. Thus, we established that paraherquamide produced a dose-dependent antagonism of levamisole-activated channels.
Paraherquamide has no inhibitory effect on channel conductance or mean open times
Figure 5
A and B, left, shows the representative effects of paraherquamide on channel conductance. It illustrates results from a patch at +75mM with three channels (G25, G35, G45) activated by 30 µM levamisole in the pipette before (Fig 5A
) and after (Fig 5B
) the addition of 1µM paraherquamide to the bath. The figure shows that the mean amplitude of each of the channels with a conductance of 27 pS, 37 pS, and 47 pS is the same before and after the addition of paraherquamide. Interestingly, Fig. 5
also shows that although the amplitude of each of the three channels is unchanged by paraherquamide, the overall frequency of the events is reduced due to the inhibitory action of paraherquamide. In all patch recordings, we found no effect of paraherquamide on the amplitudes of any of the channel subtypes.
|
Figure 5A and B
, right, illustrates representative effects of 1 µM paraherquamide (Fig. 5A
, control; Fig. 5B
, paraherquamide) on the open time distribution of levamisole activated channels. It shows the distributions of open times of G35 and G45 channels activated by 30 µM levamisole in the pipette at +75mV before and after the addition of paraherquamide to the bath. There was no decrease of in the mean open times of either of the channel subtypes. There was no evidence in this patch or in any of the G35 or G45 channels recordings that paraherquamide reduced the mean open time of the channels. The lack of effect on channel conductance or on open time is consistent with the competitive mode of action of paraherquamide (1
, 17)
.
Inhibitory effects of paraherquamide and 2-desoxoparaherquamide are selective and the % inhibition depends on the receptor subtype
We have described the selective effects of paraherquamide and 2-desoxoparaherquamide on different nAChR subtypes in muscle contraction experiments (1)
. To investigate whether these two compounds show selective effects on the G25, G35, and G45 populations at the single-channel level, we tested the inhibitory effects of 1 µM paraherquamide and 3 µM 2-desoxoparaherquamide on the Popen of the G25, G35, and G45 channels.
Figure 5C
shows the effect of 1 µM paraherquamide on the three nAChR populations (G25, G35, and G45) observed in seven patches at +75mV when activated by 30 µM levamisole. When 1 µM paraherquamide was present in the bath, ANOVA showed that the opening of the three channel populations was inhibited by different amounts (F=6.9, df=2, P=0.01). Figure 5C
shows that 1 µM paraherquamide had no effect on the G25 population but had an inhibitory effect on the G35 and G45 channels (% control value: G25 106.1±25.4%; n=5; G35 32.3±10.0%, n=5; and G45, 31.8±7.0%, n=5). A t test showed that the inhibition of the G35 and G45 channel was significant when compared to the G25 channels (P<0.05). The lack of effect on G25 channels is consistent with this channel population being the N-subtype because it is the least sensitive to paraherquamide (pKB 5.86: ref 1
). The greater effect on the G35 and G45 channels is consistent with these channels belonging to the L- and B-subtype because paraherquamide is a more potent antagonist of these two subtypes than the N-subtype (pKB, 6.6; ref 1
).
We also used 10 µM bephenium to activate the G35 and G45 channels without activating the G25 channels. Bephenium activated currents were recorded from 12 inside-out patches and the inhibitory effects on Popen of bath-applied 3 µM 2-desoxoparaherquamide on the G35 and G45 channels observed, Fig 5C
. 2-desoxoparaherquamide had a significantly greater inhibitory effect on the G45 population than the G35 population (%reduction Popen of G45: 26.7±7.1, n=5; %reduction in Popen of G35: 63.6±11.4, n=4, P<0.05). The greater effect of 2-desoxoparaherquamide on G45 channels is consistent with these channels being the B-subtype (pKB=5.8: ref 1
) and the G35 channels being the L-subtype (pKB=5.0: ref 1
).
DISCUSSION
N-, L-, and B-subtypes identified at the single-channel level on body muscle
We have observed in a muscle strip contraction assays of Ascaris suum (1)
that the cholinergic anthelmintics activate three subtypes of receptors on body muscle and that paraherquamide and 2-desoxyparaherquamide are competitive antagonists. There is the N-subtype that is preferentially activated by nicotine but not paraherquamide and 2-deoxyparaherquamide; there is the L-subtype that is preferentially activated by levamisole and antagonized by paraherquamide; and there is the B-subtype that is preferentially activated by bephenium and antagonized by paraherquamide as well as 2-desoxoparaherquamide (1)
.
In a previous study (13)
at the single channel level, comparison of the effects of nicotine and levamisole revealed that nicotine and levamisole could activate a wide range of conductance but that nicotine preferentially activated small 26 pS channels and that levamisole preferentially activated larger channels with a conductance averaging 39 pS. In this current study, we have been able dissect out the third, G45, subtype at the single channel level (Fig. 6
). Also in this study we found that levamisole activated channels over a wide conductance range and that in some patches we could identify 3 separate channels that we refer to as G25, G35, and G45 with mean conductance around 25, 35, and 45pS respectively. Levamisole activated more G35 channels than the other two subtypes. Paraherquamide had no effect on the G25 channels but antagonized the G35 and G45 channels in a manner that was consistent with the G25 channels being the N-subtype not the L- or B-subtype (1)
.
|
When we used bephenium as the agonist, it did not activate the G25 channels but activated more G45 channels than G35 channels; furthermore, 2-desoxoparaherquamide selectively inhibited the G45 channels. We found that 2-desoxoparaherquamide is a more potent antagonist of the G45 channels than the G35 pS channels and that 1 µM bephenium preferentially activated the G45 channels. These observations allowed us to identify and confirm that the G45 channel is the B-subtype and that the L-subtype is the G35 channel. 2-Desoxoparaherquamide is a more potent antagonist of the B-subtype than of the L-subtype (1)
.
In addition to separation of the subtypes by pharmacology and conductance, we found that there were significant differences in their mean open times with the N-subtype (G25) having the briefest open time, 0.6mS; the L-subtype (G35) having an intermediate open time, 0.9 ms; and the B-subtype (G45) having the longest open time,1.3 ms, when levamisole was the agonist. These experiments have allowed us to identify the three subtypes of muscle nAChR at the single-channel level and to see their pharmacology at that level. The significance of the different subtypes of nAChR is discussed in the following paragraphs.
Recognition of separate nAChR subtypes in parasitic nematodes is therapeutically significant
We have observed that L-subtype channels decrease in frequency on the body muscle of the nematode parasite Oesphagostomum dentatum (15)
and that levamisole resistant isolates of Oesophagostomum dentatum become less sensitive to levamisole (more selective for L-subtype channels) but that they remain sensitive to the N-subtype agonist methyridine (10)
. If levamisole resistance is due to the loss of the L-subtype of receptors, a useful therapeutic approach may be to use cholinergic agonists or antagonists that have selective effects against N- and B-subtypes to treat levamisole resistance. This suggests that 2-desoxoparaherquamide or, a combination of methyridine or oxantel (N-subtype selective) and bephenium may be useful for the treatment of some types of levamisole resistance.
Heterogeneity of nematode nAChRs
We have observed three separate subtypes in the parasitic nematode Ascaris suum, but what is the molecular basis of the heterogeneity? Much information on the structure and function of nematode AChRs has advanced with the study of C. elegans, a soil nematode that has been subjected to detailed genetic analysis. We are encouraged to use C. elegans as a model for the levamisole receptor of parasitic nematodes by molecular phylogenetic analysis (18)
. C. elegans belongs to clade V as do many nematode parasites like O. dentatum, and there are strong similarities between the single-channel properties of levamisole receptors of A. suum (clade III) and O. dentatum (clade V; ref 19
). In both nematodes, channel conductance subtypes between 20 – 50 pS are observed with conductance subtypes that include the G25, G35, and G45. In O. dentatum, however, we were able to separate a fourth G40 conductance subtype.
With the completion of the C. elegans genome project, the full size of the nematode AChR gene family was finally revealed. There are now a total of 27 AChR subunit genes which is bigger than in mammals (13 genes, and is the largest number of AChR subunit genes in a single species (19)
. Homology has allowed these 27 subunits to be divided into five groups named: DEG-3-like; ACR-16-like; ACR-8-like; UNC-38-like; and UNC-29-like. The UNC-29-like subunits are non-
subunits, and the remainders are mostly subunits. If all combinations of and non- subunit were capable of forming functional AChRs, there could be over 25 (5)
or 14,348,907 receptor subtypes, but this is not likely. The molecular structures of levamisole UNC-38 subunit homologues in the parasitic nematodes, Trichostrongylus colubriformis (20)
and Hemonchus contortus (21)
, are 90% identical to C. elegans, suggesting that information derived from the model nematode may be a useful guide for parasitic nematodes. With a large number of AChR subunits in C. elegans, it is not surprising that we also see heterogeneity in our experiments on parasitic nematodes.
We can see the molecular potential for the heterogeneity, but what is the functional benefit of this extensive heterogeneity? If nematodes are like vertebrates, then the physiological reasons for the heterogeneity include a necessity to have varying sensitivities to the agonist ACh, to vary desensitization, calcium permeability, and distributions in different cells (22)
. We know that expression of UNC-38, UNC-29, UNC-63, LEV-1, ACR-8, ACR 16, and ACR-13 (=LEV-8) GFP-tagged protein subunits occurs in body muscle cells of C. elegans (23
, 24
, 25
, 26
, 27)
, but details of the stoichiometric arrangements of the subunits and total number of receptor subtypes remain to be determined. In our study, we demonstrate the presence of three subtypes of AChR in muscle of parasitic nematodes, each with a different sensitivity to cholinergic anthelmintics. The identification of multiple receptor subtypes in parasites has several practical implications. Firstly, it may help understanding of how resistance to the cholinergic anthelmintics can arise: changes in the relative of proportion of each subtype (by up- or down-regulation) will alter the sensitivity of the parasite to a specific anthelmintic. Secondly, therapeutics may benefit: using combinations of compounds that activate all receptor subtypes could enhance cure rates. Finally, it is predicted that, in some instances, parasites resistant to one type of cholinomimetic will retain sensitivity to others (e.g., levamisole resistant parasites can remain sensitive to bephenium or methyridine), thus offering an alternative strategy to deal with resistance.
ACKNOWLEDGMENTS
We thank Dr. Mark Levandoski for helpful discussion on this study. We also acknowledge National Institutes of Health for financial support (R01 AI47194) to R. J. Martin.
FOOTNOTES
1 These authors contributed equally to this work. 
Received for publication May 3, 2006. Accepted for publication July 5, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  H. Qian, A. P. Robertson, J. A. Powell-Coffman, and R. J. Martin Levamisole resistance resolved at the single-channel level in Caenorhabditis elegans FASEB J, September 1, 2008; 22(9): 3247 - 3254. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A. Aloysius, M. V. S. Elipe, B. H. Arison, T. D. Faidley, B. F. Michael, T. A. Blizzard, D. R. Thompson, W. L. Shoop, and R. A. Tschirret-Guth Comparative Disposition and Metabolism of Paraherquamide in Sheep, Gerbils, and Dogs Drug Metab. Dispos., August 1, 2008; 36(8): 1659 - 1669. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
