Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
I This paper is contribution no. 850 from the HawaiiInstitute of Marine Biology. Manuscript accepted 15September 1990.
2 Department of Zoology , University of Hawaii atManoa, Honolulu, Hawaii 96822.
THE LIFE CYCLES OF MOST digenetic trematodesinclude one or more intermediate hosts andinvolve many intimate inter-relationshipswith hosts that can lead to interesting behavioral adaptations. Of particular interest isthe suggestion that some parasites alter theappearance or behavior of their intermediatehost to facilitate transmission of the parasiteto its final host (Holmes and Bethel 1972). Forexample , snails infected with the broodsacs ofLeucochloridium have tentacles that pulsate inresponse to light (Cheng 1986), and fish infected with certain heterophyid or strigeidtrematodes develop "black spots" caused byhost response to larval trematode invasion(Rothschild 1962). In both cases, it has beensuggested that the increased conspicuousnessof the intermediate hosts increased theirsusceptibility to predation by definitive host
Pacific Science (1991), vol. 45, no . 3: 263-269© 1991 by University of Hawaii Press. AIl rights reserved
Behavioral and Ecological Relationships of a Parasite and Its Hosts withina Coral Reef System 1
GRETA S. AEBy2
ABSTRACT: The life cycle of the digenetic trematode Plagioporus sp. includesan intermediate stage that encysts in the scleractinian coral Porites compressaand an adult stage that probably resides in a coral-feeding fish. Coral polypsinfected with metacercariae of Plagioporus appear as swollen nodules ranging incolor from bright pink to white and have lost their ability to retract into theircalices. The polyps' altered appearance and behavior was thought to increasetheir vulnerability to predation. This study investigated the effect of parasiteencystment on coral growth and the effect offish predation on both coral growthand on the parasites' rate of transmission. Parasitized P. compressa showed a50% reduction in growth when compared to nonparasitized P. compressa. Nosignificant differences were found in growth of corals kept in predator exclusioncages and that ofcorals left exposed to fish predation in either group, parasitizedor nonparasitized. Uncaged parasitized P. compressa showed a marked reduction in number of parasitic cysts, with the infected polyps being replaced byhealthy ones. The regeneration of healthy polyps suggests that parasite removalis beneficial to the coral, and the reduction in cyst number suggests that theparasites' rate of transmission was enhanced by exposure of infected corals tofish predation.
species. Similarly , effects of parasites on hostbehavior have also been reported. Carpenterants infected with Brachylecithum mosquensismetacercariae exhibit atypical sluggish behavior and are no longer photophobic(Carney 1969), and amphipods infected withacanthocephalan larvae have altered evasivebehavior and responses to light (Bethel andHolmes 1977). In both cases, the altered behavior was thought to increase the intermediate hosts' vulnerability to predation.
Here I present data on a digenetic trematode, Plagioporus sp ., whose presence results in the alteration of both the behavior andappearance of its intermediate host, a scleractinian coral. The parasites' metacercarialstage encysts in two species of coral, Poritescompressa and Porites lobata, from KaneoheBay, Oahu, Hawaii (Cheng and Wong 1974).The infected polyps appear as irregularlyshaped nodules on the coral, ranging in colorfrom bright pink to white . In addition, infected polyps have lost the ability to retract into their calices, which may increasetheir vulnerability to predation. Preliminary
263
264
studies suggested that the alteration of theinfected polyps resulted in corallivorous fishbeing attracted to and feeding on the infectedpolyps (unpublished data). The current studyinvestigated the effect of parasite encystmenton coral growth as well as the effect of fishpredation on both coral growth and on theparasites' rate of transmission.
Typical Life Cycle of the Parasite
The life cycle of a typical digenetic trematode is well known (Dogiel et al. 1961, Olsen1962, Eramus 1972, Smyth and Halton 1983)and usually includes the adult worm living inthe gall bladder of a fish. Eggs pass out withthe host feces and hatch into a swimmingmiracidium stage, which enters a molluscanfirst intermediate host. The parasite leaves themollusk in the cercaria stage, which penetratesits second intermediate host, Porites sp. coral.It enters the tentacles of the coral polyps andencysts as a metacercaria (Cheng and Wong1974). Completion of the life cycle occurswhen a coral-feeding fish ingests the encystedmetacercaria.
MATERIALS AND METHODS
The effect of parasitism and fish predationon the growth of host corals was examined inan experiment of factorial design: parasitized,unparasitized, protected from fish predation,and exposed to fish predation. Ten colonies ofparasitized Porites compressa and 10 coloniesof unparasitized P. compressa were haphazardly collected from patch reefs in KaneoheBay, Oahu, Hawaii . Each coral colony wascut in half with a rock saw to produce twogenetically identical specimens, with the twohalves of each parasitized coral colony havingrelatively similar levels of parasitism. Allcorals were stained with Alizarin red stain topermit measurement of linear growth (Barnes1972), from the stained skeleton to the outeredge of the corallum. The buoyant weight(Jokiel et al. 1976) of each coral colony halfwas measured at the beginning and end of the60-day experimental period to determinecolony growth rate.
PACIFIC SCIENCE, Volume 45, July 1991
To assess the effect of fish predation, onehalf of each coral colony was randomly selected for placement in a wire-mesh predatorexclusion cage . All corals were wired ontometal screens, with wire cages protectingthose designated for the fish exclusion treatment. The screens were transported to thewindward side of a fringing reef off CoconutIsland in Kaneohe Bay and secured to theedge of the reef crest with iron stakes. Thecorals were placed in an area where coralfeeding butterflyfish (Chaetodon trifasciatus,C. auriga, C. lunula, C. lineolatus, and C.ephippium) were known to reside . The experi mental site was checked weekly to ensure thatcages were free of sediment and algae. Coralswere removed after 60 days and reweighed .Linear growth was determined by cutting fivebranches from the center of each coral andmeasuring the skeletal growth in the axialplane beyond the Alizarin red stain. To assessthe parasite lode for each of the parasitized P.compressa colonies, a 3 ern x 5 em quadratwas placed on the side of a coral midwaybetween the base and top of the coral, and thenumber of cysts visible within that quadratwas recorded . A reading was taken on twoopposite sides of the coral. The unparasitizedP. compressa colonies were thoroughly examined for signs of parasite infection, and thenumber of cysts found on the colony wasrecorded.
All statistical analyses were performed withthe SAS system (SAS Institute, Inc. 1985).One coral was lost in the caged unparasitizedtreatment, resulting in a sample size of ninefor that group.
RESULTS
Parasitized P. compressa showed a significantly smaller percentage gain in weightthan unparasitized P. compressa (two-wayANOVA, P = .0007, n = 39; Table 1). Parasitized P. compressa had a mean weight gainof 11.7% (SE ± 0.88) while that of unparasitized P. compressa was 23.3% (SE ± 3.2).This reflects a 50% reduction in the growth ofparasitized P. compressa (Figure 1). A comparison of the mean linear growth also
Parasite-Host Relationships in a Coral Reef System-AEBY 265
TABLE I TABLE 2
TWO-WAYANOVA FOR MEAN WEIGHT GAIN (%) FORPARASITIZED AND UNPARASITIZED P. compressa
ONE-WAY ANOVA FOR MEAN LINEAR GROWTH OFPARASITIZED AND UNPARASITIZED P. compressa
SOURCE df MS* F SOURCE df MS* F
Parasites I 0.2101 13.78 P = .0007 Parasites 1 14.7622 5.85 P = .0204Caging I 0.0000 0.00 P = .9668 Error 38 2.5218Paras x Cage I 0.0127 0.84 P = .3662Error 35 0.0152 *Indicates mean square.
• Indicates mean square.
pa ras itized uncagedcaged
c:'iiiCl_ 20
s:Cl'iii::c:1tIell 10E
30
caged uncagedunparasitized
30
c:'iiiCl 20
ECl
.~
c::ll 10
E
showed similar significant differences between the parasitized and unparasitized coralcolonies (one-way AN OVA, P = .0204; Figure 2, Table 2).
A Wilcoxon signed-rank test showed nosignificant differences in growth between
FIGURE3. Comparison of mean weight gain (%) forcaged and uncaged treatments of parasit ized P. compressa(Wilcoxon signed-rank test, P = .2867, n.s., n = (9) andunparasitized P. compressa (Wilcoxon signed-rank test,P = .8405, n.s., n = 20). Bars indicate standard error.
unparas itized
unpar asili zed
pa rasitized
pa ras itized
30
c:'iiiCl 20
ECl'iii::c::ll 10
E
FIGURE 1. Comparison of mean weight gain (%) forparasitized and unparasitized P. compressa. Bars indicatestanda rd erro r. (Two-way ANOVA, P = .0007, n = 39.)
FIGURE2. Comparison of mean linear growth forpara sitized and unparasitized P. compressa. Bars indicatestandard error. (One-way ANOVA, P = .0204, n = 40.)
266 PACIFIC SCIENCE, Volume 45, July 1991
caged uncagedun parasitized
FIGURE 4. Mean linear growth of caged and uncagedtreatments for parasitized and unparasitized P. compressa. Bars indicate standard error.
caged and uncaged P. compressa in eithergroup, parasitized or unparasitized (Figure 3).Parasitized P. compressa within predatorexclusion cages had a mean weight gain of10.7% (SE ± 1.5) compared with uncagedparasitized P. compressa, which had a meanweight gain of 12.8% (SE ± 0.92) (T =0.2867, n.s.). Unparasitized P. compressawithin predator-exclusion cages had a meanweight gain of 25.0% (SE ± 5.5) comparedwith uncaged unparasitized P. compressa,which had a mean weight gain of 21.9%(SE ± 3.7) (T = 0.8405, n.s.). A comparisonof the mean linear growth also showed nosignificant differences between caged and uncaged P . compressa (Figure 4).
At the end of the 60-day experimentalperiod, uncaged parasitized P. compressa hadsignificantly fewer parasitic cysts than the P.compressa protected from fish predation(Wilcoxon signed rank test, T = 0.0075).Caged parasitized P . compressa had a meannumber of parasites of 34.3 ± 22.9 cysts/30em2 as compared with the uncaged parasitizedP. compressa, which had a mean number ofparasites of9.8 ± 5.2 cysts/30 em? (Figure 5).The unparasitized P. compressa showed signsof becoming newly infected in both caged anduncaged treatments (x = 2.1 parasites percoral colony, SD ± 2.3).
parasitized uncagedcaged
E'6E:;5so 40,...~ 3c:
c: 2
'"QlE,
FIGURE 5. Mean number of parasites for caged anduncaged treatments of parasitized P. compressa. Barsindicate standard error. (Wilcoxon signed-rank test,P = .0075, n = 20.)
DISCUSSION
A parasite, by definition, is an organismthat lives in intimate association with and hasan unfavorable impact on another organism,its host (Minchella 1985). Porites compressainfected with Plagioporus sp. had an overallreduction in growth rate consistent with thisdefinition ofa parasitic infection. Some stagesof parasites derive their food directly from thehost, but after encystment the metacercariaeof Plagioporus sp. place no metabolic demands on the coral host. However, Plagioporus affects P. compressa in a number ofother ways, which suggests that there is somecost to the polyps and that their contributionto the growth of the coral colony may bereduced. First, infected polyps appear swollen,and they are unable to retract adequately into
uncagedcaged
Qi 20.cE:::lc:c:~ 10
E
E 40ooMCilQl
~ 30
~
'"Co
Parasite- Host Relat ionships in a Coral Reef System-AEBY 267
their calices. Second, infected polyps have areduced number of zooxanthellae; becausecoral growth is enhanced by the presence oftheir symbiotic zooxanthellae (Muscatine1973), the lack of zooxa nthellae in infectedpolyps may have contributed to the reducedgrowth. Last , infected polyps pro duce small ,abnormal skeletal protrusions; Cheng andWong (1974), who found significantly lessionic calcium in the polyps of parasitized P.compressa as compared with unp arasitized P.compressa, suggested that the presence ofPlagioporus metacercariae in a polyp resultsin the disruption of normal calcium deposition by polyp epidermal cells. It should alsobe noted that the coral s used in this experiment had high levels of infection (Figu re 5),which may have resulted in the large reductionin growth exhibited by parasitized P. compressa. It is hypothesized that a correlationexists between level of parasite infection anddegree of reduced coral growth, but furtherinvestigation is needed to verify this .
Different genotypes of P. compressa mayhave various susceptibilities to parasitism.However, the unparasitized corals in thisstudy showed signs ofinfection after 2 monthson a reef where Plagioporus occurs, suggestingthat the lack of previous infection of unparasitized P. compressa was due to lack of exposure to the parasite rat her than an inherentimmunity .
Presence of a larval parasite often increasesthe rate of predation on the intermediatehost by the definitive host (Ho ogenboom andDijkstra 1987, Quinn et al. 1987, Godin andSproul 1988), thereby increasing the opportuniti es for the parasite to complete its lifecycle. It follows that selection would favor theparasite whose encystment results in an increase in the vulnerability of the intermedia tehost to predation by the definitive host. Thiswould be particularly important where theparasites' encysted metacercarial stage has alimited period of viability, with senescenceoccurring if it is not ingested by the definitivehost (Eramus 1972). The parasite would thenhave only a certain amount of time in whichto complete its life cycle. This seems to beconsistent with Plagioporus, which does havea finite metacercarial stage that usually
seneces within a few months (unpublisheddata). Coral-feeding butterflyfish are attracted to and readily feed on infected corals.In addition, parasitized P. compressa exposedto fish predation did experience a dramaticreduction in the number of cysts, suggestingan enhancement of the parasites' rate oftransmission.
There were no significant differences ingrowth found between P. compressa exposedto fish predat ion and those protected fromfish predation. Th is was true for both parasitized P. compressa and unparasitized P.compressa. This is in contrast with otherstudies that have shown that corallivorousfish can limit the growth and distribution ofcorals (Neudecker 1979, Cox 1986). However,the previous studies used species ofcoral otherthan Porites. Porites has a lower energy content than other species of coral found on theexperimental reef (Tricas 1985), and it is notthe preferred food of most butterflyfishes(Reese 1977, Tricas 1985). Therefore, the levelof fish pred ation on P. compressa may nothave been large enough to significantly affectits growth. This is a reasonable explanationfor unparasitized P. compressa; however, thereduction in number of parasitic cysts onparasitized P. compressa indicates that predation was probably occurring. Why then wasthe growth rate of parasitized P. compressanot reduced? The effects of corallivores areprobably similar to those of herbivorousgrazers, where only partial predation occurswithout complete loss of the coral polyp (Cox1983). I observed that corals regeneratedhealth y polyps after removal of the encystedparasite. The negative effect of fish predationmay have been offset by the positive effect ofthe parasite removal.
One might question why fish would evolveto feed on infected polyps, thereby becomingparasitized themselves, instead of avoidingthe cysts. I can only speculate about this, butthere are several hypothetical explanations:
1. The parasite residing in the fish mayhave adopted the " prudent parasite" strategy(Holmes 1983) in which the parasite producesminimal damage in the host. There wouldthen be a lack of selective pressure on the fishto avoid the infected polyps .
268
2. There may be a positive selective pressure on the fish to maximize feeding efficiencyby eating the infected polyps that are nolonger able to retract into their cal ices.
3. It may also be a combination of theabove two hypotheses in which there is a positive selection to ma ximize feeding efficiencyby feeding on the infected polyps combinedwith minimal parasitic damage to the host,crea ting a lack of pres sure to avoid the infected polyps.
In summary, the presence of the parasitePlagioporus has produced an alteration inboth the appearance and the behavior of itsintermediate host, P. compressa. These alterations have resulted in increased predation onP. compressa, facilitating the tr ansmission ofthe parasite to its probable final host. Thegrowth of P. compressa, ho wever , was notsignificantly affected by exposure to fish predation. Encystment of Plagioporus met acercari ae was found to have a negat ive impact onthe growth of P. compressa.
LITERATURE CITED
BARNES, D . J . 1972. The struc ture and formation of growth ridge s in sclerac tinian coralskeletons . Proc. R. Soc. London 182: 331350.
BETHEL, W. M ., and J. C. HOLMES. 1977. Increased vulnera bility of amphipods to predation owing to altered behavior inducedby larval acanthocephalan s. Can. J. Zool,55: I 10- I 15.
CARNEY, W. P. 1969. Beha vioral and morphological changes in carpenter ants harboring dicrocoeliid metacercariae. Am .MidI. Nat. 82(2): 605-61 I.
CHENG, T. C. 1986. General parasitology.Academic Press, Orlando, Florida.
CHENG, T . c..and A. K. WONG. 1974. Ch emical, histochemical and histopathologicalstudies on corals, Por ites spp. , parasitizedby trematode metacercari ae. J . Invertebr.Pathoi. 23 :303-317.
Cox, E. F . 1983. Aspects of coral1ivory byChaetodon unimaculatus in Kaneohe Bay,Oahu. M.S. thesis, Universi ty of Hawaii .
PACIFIC SCIENCE, Volume 45, July 1991
- - - . 1986. The effects of a select ive corallivore on growth rates and competition forspace between two species of Hawaiiancorals. J . Exp. Mar. BioI. Ecoi. 101 : 161174.
DOGIEL, V. A., G . K . PETRUSHEVSKI, andY. I. POLYANSKI. 1961. Pari sito logy offishes. Oliver and Boyd , Edinburgh andLondon.
ERAMUS, D . A. 1972. The bio logy of trematodes. Crane, Russak and Co. , New York.
GODIN, J ., and C. SPROUL. 1988. Risk takingin parasitized stickle-backs under threat ofpredation: Effects of energetic need andfood avail ability. Can. J. Zool. 66 :23602367.
HOLMES, J . C. 1983. Evolutionary relationships between parasitic helminths and theirhosts. Pages 16I- 185 in D . F . Futuyma andM . Slatkin, eds. Coevolution. Sinauer Press,Sunderland, Massachusett s.
HOLMES, J . c., and W . M . BETHEL. 1972.Modifica tion s of intermediate host beha vior by parasites. Pages 123- 149 in E. V.Canning and C. A. Wrigh t, eds. Behavior alaspects of parasite transmission. AcademicPress, New York .
HOOGENBOOM, I., and C. DIJKSTRA. 1987.Sarcocystis cernae : A parasite incre asingthe risk of predation of its int erm ediat ehost, Microtus arvalis. Oecologia (Berlin)74 : 86- 92.
JOKIEL, P. L., J. E. MARAGos, and L.FRANZISKET. 1976. Measurement of skeletal growth in scleractinian corals bybuoyant weight technique. In D . R.Stoddart and R . E. Johannes. Handbook ofcoral reef research methods. UNESCOMonogr. Oceanogr. Methodoi.
MINCHELLA, D. J. 1985. Host life-historyvaria tion in response to parasitism. Par asitology 90 :205-216.
MUSCATINE, L. 1973. Nutrition of corals.Pages 98- I 15 in O. A. Jones and R. Endean,eds. Biology and geology ofcoral reefs, Vol.II , Biology I. Academic Pres s, New York.
NEUDECKER, S. 1979. Effects of grazing andbrowsing fishes on the zonation of corals inGuam . Ecology 60: 666-672.
OLSEN, O. W . 1962. Animal parasites: Their
Parasite-Host Relationships in a Coral Reef System-AEBY 269
biology and life cycles. Burgess Publ.,Minneapolis, Minnesota.
QUINN, S., R. BROOKS, and R. CAWTHORN.1987. Effects of the protozoan parasiteSarcocystis rauschorum on open-field behavior in its intermediate vertebrate host,Dicrostonyx richardsoni. J. Parasitol. 73(2) :265-271.
REESE, E. S. 1977. Coevolution of corals andcoral feeding fishes of family Chaetodontidae. Proc. 3rd Int. Coral Reef Symp .I: 267-274.
ROTHSCHILD, M. 1962. Changes in behavior
in the intermediate hosts of trematodes.Nature (London) 193: 1312-1313.
SAS INSTITUTE, INC. 1985. User's guide: Statistics . Version 5. SAS Institute, Inc. , Cary,North Carolina.
SMYTH, J. D. , and D. W. HALTON. 1983. Thephysiology of trematodes. 2d ed. CambridgeUniversity Press , Cambridge.
TRICAS, T. C. 1985. The economics of foraging in coral-feeding butterflyfishes of Hawaii. Proc. 5th Int. Coral Reef Congr.,Tahiti 5:409-414.