Waterborne and Surface-Associated Carbohydrates as Settlement Cues for Larvae of the Specialist Marine Herbivore Alderia modesta

10 Pages·1999·4.6 MB·English

by  P J Krug

Reference: Biol. Bull. 197: 94-103. (August 1999)
Waterborne and Surface-Associated Carbohydrates as
Settlement Cues for Larvae of the Marine
Specialist
Herbivore Alderia modesta
PATRICK J. KRUG1 * AND ADRIANA E. MANZI2
' Department ofBiology, University1 ofCalifornia, Box 95J606, Los Angeles, California 90095-]606;
and 2 Cytel Corp., 9393 Towne Center Drive, San Diego, California 92121-3016
Abstract. Larvae ofthe specialist marine herbivore Alde- Introduction
ria modesta (Opisthobranchia: Ascoglossa) metamorphose
in response to a chemical settlement cue from the alga Most marine invertebrate species produce free-swimming
Vaucheria longicaulis, the obligate adult prey. Bioactivity larvae that disperse in the plankton until becoming compe-
coeluted with both high and low molecular weight carbo- tent to settle to the bottom and metamorphose into the adult
hydrates in solution, and with insoluble high molecular form (Grahame and Branch, 1985; Levin and Bridges,
weight carbohydrates associated with the algal cell wall. 1995). Larval recruitment plays a critical role in benthic
Larvae metamorphosed in response to water conditioned by marine ecosystems, structuring communities and regulating
V. longicaulis, as well as to frozen and homogenized algal population dynamics (Grosberg, 1982; Roughgarden et al.,
tissue. The inducer was efficiently extracted from the algae 1988; Underwood and Fairweather, 1989). Microscopic lar-
with boiling water, but after all soluble activity was ex- vae are generally viewed as passive particles transported by
tracted, residual tissue still induced larval settlement. Etha- flow to the benthos (Eckman, 1983. 1990; Butman, 1987).
nol precipitation of a boiled-water extract followed by gel Following hydrodynamic delivery of larvae to the bottom,
recruitment can be divided into settlement and metamor-
filtration chromatography showed that the precipitate con-
tained carbohydrates of >100,000 Da molecular weight, phosis (Chia and Koss, 1988; Pawlik, 1992). Settlement is
while the supernatant contained only low molecular weight characterized by active behaviors with which larvae explore
baclsaesrob4coi%haytdNerdaatOweiHsth(ex<tth2re,a0cct0a0rwbaDosah)yc;dhrraiontmebaoptteohagkrc.aapsAhensedaalqilunea7ocutMsiv-iiutnryseowalauts-o bts1the9re9ca3top)amh.yi(sLnLiagecraTvrloaeuesarunmnsdaepueycnxhdreeeamjdneicdcitanlaBtohscueuhrbagwsretaatrtc,ateterer1ica9so8nt8lid;cusrmeRnosofud(mrBpeioutgtseumnwetaizinmalmeetitsnuaagllb..,-,,
AycietlidviatybiwoaacstivneothiagfhfemcotledecublyarprwoetiegihntascearKbohoyrdrmaitledpeaacki.d r1e9s8p8;onBduttomsaunrfaancde-Garsassoscliea,te1d99c2u)e.sAalntdercnaotmimveilty,tolamrevtaaemmoary-
hydrolysis, but was significantly decreased by periodate phosis, an irreversible developmental transformation into
treatment. The results indicate that larvae of A. modesta the adult stage ofthe organism (Burke, 1983; Pawlik et al.,
metamorphose in response to both water-soluble and sur- 1991; Roberts etal., 1991; Pawlik, 1992). Larvae are capa-
face-associated carbohydrates of V. longicaulis, and that the ble of fine-scale discrimination among substrata both in the
soluble cue exists as both high and low molecular weight laboratory and in the field (Keough and Downes, 1982;
isoforms. Raimondi, 1988).
Recent studies have demonstrated that both surface-asso-
ciated and water-soluble chemical cues can trigger larval
behavioral responses that greatly increase rates of settle-
*ReTcoeivwehdo3mMacrocrrhes1p9o9n9d;eanccceepstheodul1dJubnee a1d9d9r9e.ssed. E-mail: pkrug@ ment and metamorphosis. Still-waterlaboratory assays have
biology.ucla.edu demonstratedthe importanceofsurface-associatedchemical
Abbreviations: BVE = boiled Vaiicheriu extract. cues for inducing larval metamorphosis of barnacles (Maki
94
CARBOHYDRATE SETTLEMENT CUES 95
et til., 1990), bryozoans (Hurlbut, 1991), corals (Morse et were grown undercontinuous lighting in the laboratory, and
al., 1988), gastropods (Morse et nl., 1984), and polychaetes blades of algae were pulled free of the sediment base and
(Kirchman et al.. 1982). Hydrodynamic conditions and the rinsed in seawater before use in assays. Adult specimens of
presence ofa surface cue associated with adult conspecifics A. modesta were maintained in petri dishes under 1 cm of
had an interactive effect on settlement of larvae of the seawater. and lecithotrophic egg masses were harvested
reef-building polychaete Phragmatopoma califomica in daily for3 days. Egg massesfromeachday werepooledand
flow (Pawlik et al., 1991). Waterborne chemical cues also maintained in 0.45 jam-filtered seawater (FSW); water was
affect larval settlement processes. Soluble cues secreted by changed every other day until hatching. Upon hatching,
the adult prey organisms induced settlement and metamor- larvae were maintained in FSW for 2 days, to allow spon-
phosis in the opisthobranchs Pliestilla sibogae andAdalaria taneous metamorphosis to occur in cue-independent larvae
proximo (Hadfield and Scheuer, 1985; Lambert and Todd, (Krug. 1998a). The remaining larvae were then subsampled
1994). Larvae of the oyster Crassostrea virginica showed for use in the bioassay. Foreach experimental treatment, 15
dramatic behavioral responses to achemical cue secreted by larvae were added to each of3 replicate dishes containing 4
adult conspecifics, increasing settlement in both still and ml FSW. After 2 days, larvae were scored formetamorpho-
moving water (Tamburri et al., 1992; Turner et al., 1994; sis. Each experiment included a FSW-only treatment as a
Tamburri et al.. 1996). However, despite decades of re- negative control and live V. longicaulisasapositivecontrol.
search into the nature of larval chemical settlement cues, The percentage of metamorphosis for each replicate was
relatively little is known about the molecules that regulate arcsine transformed, and treatments were compared using a
this crucial aspect ofthe life history ofmost benthic marine 1-way ANOVA. Unplanned comparisons of means were
invertebrates. done using the Scheffe procedure (Day and Quinn, 1989).
A recent study of a population of the opisthobranch
mollusc Alderia modesta revealed several unusual features
Secretion ofsettlement cue
that make A. modesta an ideal experimental system for
investigating larval life history and settlement processes An experiment was designed to determine whether the
(Krug, 1998a, b). A. modesta is an ascoglossan found in Vaucheria-denved settlement cue was surface-associated or
temperate estuaries in association with its obligate food secreted by the algae. Small patches (1 cm2) of V. longi-
source, the yellow-green alga Vaucheria longicaulis (Xan- caulis were cut from a growing mat and left attached to the
thophyta: Xanthophyceae) (Hartog and Swennen, 1952; sediment base. Conditioned seawater (CSW) was made by
Hartog, 1959; Trowbridge, 1993). In southern California,A. placing a patch in 4 ml FSW for either 3 h or 24 h, after
modesta exhibits a reproductive polymorphism that is ex- which the CSW was filtered through cotton and placed in a
CSW
tremely rare among marine invertebrates; study populations sterile petri dish; larvae were added directly to the for
contain specimens that produce planktotrophic larvae and the bioassay. Conditioned fresh water (CFW) was made by
other individuals that produce lecithotrophic larvae (Krug, placing patchesofV. longicaulis in4 ml deionized waterfor
1998b). Most lecithotrophic spawn masses contain a mix- 24 h. The CFW was filtered through cotton, dried on a
ture of sibling larvae, some of which metamorphose spon- rotary evaporator, and resuspended in an equivalent volume
taneously within 2 days ofhatching; the remaining veligers of FSW for use in the bioassay. The negative control was
delay metamorphosis until encountering a chemical cue FSW aged 24 h and filtered through cotton in parallel with
derived exclusively from the adult host alga V. longicaulis treatements; the positive control was live V. longicaulis
(Krug, 1998a). The present work used a bioassay for larval tissue.
metamorphosis to determine whether the inductive activity To determine whether Vaucheria longicaulis must be
was soluble or surface-associated in nature, and for bioas- alive to trigger metamorphosis, pieces of the algae were
say-guided isolation ofactive fractions as a preliminary step frozen at -20C for3 days. Frozen patches were thawed by
in purifying the settlement cue. immersion in FSW at room temperature for 1 h prior to use
in the bioassay. To determine whether algal tissue must be
Materials and Methods physically intact, blades of live V. longicaulis were pulled
free of a 2 cm2 sediment base and washed in FSW. The
Collection oforganisms and lan'al bioassay algae was manually homogenized in 10 ml deionized water
Alderia modesta (Loven, 1844) and Vaucheria longi- for20 min. and the suspension sonicatedforanother 10min.
caulis were collected from mudflats in the Kendall-Frost The homogenate was centrifuged (10 min, 2000 RPM) and
Marine Reserve and Northern Wildlife Preserve, and in the the supernatant removed. The soluble homogenate was as-
San Diego River Flood Control Channel, San Diego, Cali- sayed by adding 200 /Ltl (high concentration) or 30 ju.1 (low
fornia, U.S.A. All algae used in this study conformed to concentration) aliquots to4 ml FSW foruse in the bioassay.
published descriptions of V. longicaulis from California The negative control was FSW, and the positive control was
(Abbott and Hollenberg, 1976). Patches of V. longicaulis live intact V. longicaulis tissue.
96 P. J. KRUG AND A E. MANZI
Sequential extraction with boiling water min standards. Bioactivity was determined using the larval
settlement bioassay.
Four 20 X 20 cm mats of Vaucheria longicaulis attached Another 3 ml ofBVE was precipitated with 6 volumes of
to the natural sediment base were field collected (March ethanol overnight, and the supernatant and precipitated ma-
1997) and grown in the laboratory under continuous light- terial were separated as before. The carbohydrate elution
ing, moistened daily with 50% seawater. After 2 weeks profiles of both the supernatant and precipitate fractions
algal blades had grown 1-2 cm in height, and were har- were determined using a gel filtration column (90 cm X 1
vested by cutting with dissecting scissors just above the cm) ofSephacryl S-200 resin (Pharmacia Co.). The column
sediment base. The V. longicaulis tissue (1.34g wetweight) was calibrated for molecular weight using Blue Dextran to
was placed in a beaker containing 50 ml deionized water determine the void volume (V ) and glucose to determine
and boiled for 10 min. The solution of boiled Vaucheria the included volume (V,) for small molecules; size stan-
extract (BVE) was filtered through 100 /j,m Nitex mesh to dards were detected in fractions after collection visually
remove Vaucheria residue, and then through a 0.45 ;u,m (Blue Dextran) or by the phenol-sulphuric colorimetric as-
filtermembrane. The Vaucheria residue was collectedoffof say (glucose). The supernatant residue was dissolved in a
the mesh filter, put in 50 ml of fresh deionized water, and minimal volume and loaded onto the column, eluting with
again boiled for 10 min to generate a second extract. This MilliQ-purified water at a flow rate of6 ml/h and collecting
process wasrepeated fourmore times, yielding a total ofsix 0.5 ml fractions. Aliquots were taken fromeach fraction and
sequential boiling water extracts. The Vaucheria residue analyzed for carbohydrate content by the phenol-sulphuric
remaining after the sixth extraction was collected from the colorimetric assay and for protein content by the BCA
filter; this residue was yellow-brown in coloration but the assay; the detection limit for both colorimetric assays was
bwlaasdeassswaeyreedsbtiyllapdhdyisnicgalaly50intjaidct.alEiqaucoht otfo t4hemlsixFeSxWtrapcetsr 0pr.o5fil/ej,,g/fmrla.ctBioanssedrepornesetnhteinrgeseuvletriyng8cmalrbowheyrderaptoeoleeldutainodn
replicate assay dish. Pieces of live V. longicaulis were lyophilized to give 5 total fractions spanning the void vol-
assayed as a positive control, and equivalently sized pieces ume and included volume. Each pooled fraction was dis-
ofthe V. longicaulis residue remaining after the six sequen- solved in water and 150 ^il aliquots were bioassayed. The
tial extractions were also assayed. precipitated fraction was chromatographed in an identical
manner and fractions were collected, assayed for carbohy-
Biochemical characterization ofboiled Vaucheria drate content, and pooled to give five total fractions. Each
longicaulis extract (BVE) wpoeorleedbiforaascstaiyoend.waAs dpiossistoilvveedcoinntrwoalteursianngdl7i5vejuVlaualcihqeuroitas
The initial extract made by boiling Vaucheria longicaulis longicaulis induced 84 10% metamorphosis, while a
for 10 min (described above) was subjected to preliminary negative control using FSW gave 4 4% background
biochemical characterization. Six volumes of ethanol were metamorphosis.
added to 1 ml of BVE and the solution was precipitated
overnight at 4C. The precipitate was pelleted by centrifu- Sequential extraction ofVaucheria longicaulis with
gation, the supernatant removed, and the precipitate washed solvents ofincreasing polarity
with ethanol and repelleted. The supernatant and wash eth-
anol were combined and dried on a rotary evaporator. The To determine whether macromolecules associated with
precipitate and supernatant residue were individually resus- the algal cell wall were bioactive, Vaucheria longicaulis
pended in 1 ml of MilliQ-purified water, such that the was sequentially extracted with solvents of increasing po-
material in each fraction was present in solution at the same larity and harshness to extract molecules of increasing mo-
concentration as in the original extract. lecularweight. Lyophilized Vaucheria longicaulis (500 mg)
Aliquots (100 ju,l) of the initial BVE and of the resus- was homogenized into a fine powder and extracted with
pended solutions ofsupernatant and precipitate were used in 80% aqueous ethanol (50 ml, 7 h, 75C), cold water(50 ml,
subsequent assays to determine the dry weight, carbohy- 4 d, 20C), hot water (50 ml, 24 h, 65C), and 4% sodium
drate content, protein content, and bioactivity ofeach sam- hydroxide (50 ml. 24 h, 20C) (Cleare and Percival, 1972).
ple. Lyophilized aliquots were weighed to determine dry The ethanol extract was partitioned into a water-soluble
mass. Carbohydrate content was determined for duplicate fraction and a water-insoluble organic fraction. The cold
aliquots from each sample using the phenol-sulphuric col- and hot water extracts were precipitated with ethanol as
orimetric assay (DuBois et al.. 1956). Measurements were before to generate supernatant and precipitate fractions for
calibrated to a standard curve generated with known con- each extract. Aliquots corresponding to 250 /ng dry weight
centrations of glucose. Protein content was determined us- were taken from the water-soluble ethanol extract and from
ing the BCA colorimetric assay (Pierce Co.) calibrated to a the cold and hot water supernatant and precipitate fractions
standard curve generated with commercially supplied albu- and were assayed directly for bioactivity. An aliquot ofthe
CARBOHYDRATE SETTLEMENT CUES 97
organic-soluble material from the ethanol extract was dis- ples were compared using an unpaired two-tailed t-test on
solved in methanol, added to a dry culture dish, the solvent arcsine-transformed percentages for each of the three treat-
evaporated, and 4 ml of FSW added prior to the bioassay. ments, as different quantities of BVE were treated and
The 4% NaOH extract was exhaustively dialyzed against bioassayed in each case.
MilliQ-purified waterand lyophilized, giving a dry material
(44 mg) that was completely insoluble in water but dis- Results
M
solved readily in 7 Murea. The S-200 Sephacryl column Secreted and surface-associatedforms ofthe larval
was equilibrated in 7 ureaand calibrated for V,, and V, as settlement cue
before. A portion ofthe 4% NaOH extract was dissolved in
M
a minimal volume of 7 urea and loaded onto the S-200 Previous work had demonstrated that Alderia modesta
column. The sample was chromatographed and fractions larvae metamorphosed specifically in response to living
were collected and assayed exactly as before, except the tissue of Vauchcrid Innxicunlis (Krug, 1998a). The initial
M
column was eluted with 7 urea. Fractions comprising the aim of the present study was to determine whether the
high molecular weight carbohydrate peak were pooled and settlement cue was secreted into seawater by living algae,
dialyzed exhaustively against water using 10.000 molecular and whetherdead or homogenized algal tissue could induce
weight cutoffdialysis tubing. The dialysate was reduced to settlement. Waterpreviously conditioned by the presence of
a volume of 1 ml on arotary evaporatorand 100 ;ul aliquots V. longicaiilis was as active in promoting metamorphosis as
were bioassayed. was the living algae (Fig. 1A, and results of a 1-way
ANOVA: df = 4. 22; F = 32.73; P < 0.0001). The
Treatment ofEVE with proteinase K, sodium periodate,
and mild acid hydrolysis
Chemical and enzymatic treatments were performed to 120,
A
determine the biochemical nature of the settlement cue. 100
solution ofsodium periodate (0.37 M, 100 jal) was4aCdded to r1
1.0 ml ofBVE, and the solution was incubated at in the
BO
dark (Hassid and Abraham, 1957). The reaction was
quenched after 24 h by the addition of excess glycerol (20 40.
;u.l). As acontrol, 1.0mlofBVEwas incubated at4C in the
dark for24 h, afterwhichexcess glycerol (20 /il) was added livclanchena CSW(3hl CSW(24h) CFW(24h)
followed immediately by periodate as in the treated sample.
Both samples were incubated for 1 h to allow the consump-
tion of excess periodate, and were then dialyzed exhaus-
tively against deionized waterforone week. Both treatment
and control samples were lyophilized, dissolved in 300 p.\
FSW, and 100 jul aliquots usedas replicate treatments in the
larval settlement bioassay.
Proteinase K (600 /j,g) was added to a sample of BVE
(300 /ill) which had been adjusted to pH 7.8 and incubated
at 50C for 24 h. The proteinase was then inactivated by
hperaotteiinngasate 1to0B0VCEfoirmm1e5dmiiant.elAy pcroinotrrotlohweaastidnognaetb1y0a0dCdifnogr liveIauciiena dIetaitdiL-ilnitta'cnta (hhoimgohgceonnaete) h(olmoowgceonnaet)e FSW
15 min. Samples were split into three replicate 100 pil Figure 1. Inductionoflarval metamorphosisby live Vaiicheria longi-
aliquots and tested in the larval bioassay. A mild acid caiilis, deadtissue,andconditioned water. Percentagesoflarval metamor-
hM/jy.1d)rTtoFolAyBs.iVsETwha(e4s0sp0oelfruiftl)oiortnmoeawdcahbsiyevhaeedadatienfdginacalotncco1en0ncte0rnCattreafdtoirTonE75Aof(m01i..n51 wcpuohennoprtsleriaosnclnoaaermendpdagcrifoveimledptnaerrwaeiisdsthmosneeasaa.nw1saL-tiw+evareyS(VDAF.NS(lWnOon)Vg=iAacs3,a)ia;ilwnaiierstgchtasitiasinsveupe-eotscrwtoaa-nntshsrofocuolsreSmAcde.hdeaSspfeefcarercepettoneisstoitatngifevooserf
(Lahaye and Ray, 1996) and dried under vacuum to remove larval settlement cue by living V. longicaulis. Means are percentages of
TFA. As control for the presence of residual TFA salts. metamorphosis induced by exposure to Wwc/ima-conditioned seawater
BVE (400 jul) was heated in parallel at 100C for 15 min. (CSW) or conditioned fresh water (CFW). Duration ofconditioning pro-
atonddrcyoinncgenutnrdaetredvTacFuAumw.asSaamdpdleedstwoeBrVeEdiismsmoeldveidatienly30p0rijoiirl cslien<gs;nsiicfitsiicugalinivtsle.ynP(irnPevp<iaorue0sn.lt0yh0e1sfer).so-zBe.MneIaanndnduscttnhioavtewjeeodfifnVeecadtucbohyfeadreihaaodrtiiozsrosnuhte,aolmoorlgianeleniidqiuzfoeftdesreoV.df
FSW, and 100 ju.1 aliquots used as replicate treatments in the homogenized algal tissue, were assayed for inductive effect. Means not
bioassay. Differences between treatment and control sam- joined by a horizontal line differed significantly (P < 0.0?l.
98 P. J. KRUG AND A. E. MANZI
conditioning process occurred rapidly in the laboratory,
such that water conditioned for 3 h induced the same level
of metamorphosis as water conditioned for 24 h. Fresh
waterwas alsoconditioned by the presence of V. longicaulis
(Fig. 1A). There was no statistical difference between the
level ofmetamorphosis induced by the living algae and any
of the conditioned water treatments, all of which differed
significantly from the seawater-only control (Scheffe test.
P < 0.001).
Vaucheria longicaulis tissue that was frozen and thawed
induced significant larval metamorphosis, indicating that
the algae does nothave tobe alive totriggersettlement (Fig.
IB. and results ofa 1-way ANOVA: df = 4, 16; F = 61.55;
P < 0.0001). Homogenates ofalgal tissue were also active,
confirming that V. longicaulis tissue does not have to be
alive or intact to induce metamorphosis (Fig. IB). Signifi-
cantly higher levels of metamorphosis were induced by
frozen V. longicaulis and the higher concentration of tissue
homogenate than by the negative control (Scheffe test, P <
0.05). The lower concentration of homogenate did not in-
duce significantly more metamorphosis than the negative
control, indicating that the larvae may be dose-responsive to
preparations of the cue; dilution experiments with condi-
tioned seawater support this conclusion (data not shown).
When Vaucheria longicaulis was extracted with boiling
water, the resulting aqueousextract was as active as positive
controls when assayed at an 80-fold dilution (Fig. 2, and
results of a 1-way ANOVA: df = 8, 18; F = 20.45; P <
0.0001). Conditioned seawater had no effect at such a
dilution, indicating that boiling water extracted the settle-
mentcue more efficiently than did the conditioning process.
When the V. longicaulis tissue was re-extracted with boiling
water for a second time, the resulting extract induced a low
level of metamorphosis, but not significantly more than the
negative control when assayed at an 80-fold dilution (Fig.
%
metamorphosis
Ml
CARBOHYDRATE SETTLEMENT CUES 99
Table I
Comparative dry weight, protein content, carbohydrate content, and
bioactivity (SD) of100 /J aliquots ofa standardsolution ofboiled
Vaucheria extract (BVE) andtheprecipitate andsupernatant resulting
from ethanol treatmentofBVE. Theprecipitateandsupernatant were
dissolvedin thestarting volume ofextractandaliquots were removed
forchemicalassays (n = 2) andbioassays (n = 3)
100 P. J. KRUG AND A. E. MANZI
metamorphosis nudibranch Adalaria proxima metamorphosed in seawater
conditioned by the preferred adult prey, the bryozoan Elec-
- tro pilosa (Lambert and Todd. 1994). However, metamor-
liveVaucheriatwiga<it<ln
phosis ofA. proxima larvae could only be induced by live
EtOHforgamcl - colonies of E. pilosa and not by dead colonies or homoge-
EtOH(aqueous) - nizedextracts (Todd ctai. 1991; Lambert and Todd. 1994).
Coldwatersupemalani In contrast, dead and homogenized V. longicaulis tissue
induced metamorphosis in A. modesta.
Coldwaterprecipitati Secreted settlement cues are also involved in gregarious
settlement of some species. Larvae of the sand dollars
Hotwaterprecipitate - Dendraster excentricus and Echinarachinus parma meta-
4,,NaOH(resolubilized) - morphosed in response to sand beds and seawater condi-
tioned by the presence of adult conspecifics (Burke, 1984;
Filteredseawater - Pearce and Scheibling, 1990). The most detailed studies on
the effects of a secreted chemical settlement cue have fo-
Figure 5. Bioactivity ofsequential extracts of Vaucheria longicuulis.
Lyuphilized Vaucheria powder was sequentially extracted with aqueous cused on the oyster Crassostrea virginica. Larvae altered
ethanol, cold water, hot water, and 4% NaOH. The ethanol extract was their swimming speed and turning rate in response to small
partitioned intoaqueousandorganic fractions.Coldandhotwaterextracts basic peptides secreted by adult conspecifics, significantly
wereprecipitatedwithethanoltoyieldsupernatantandprecipitatefractions increasing settlement in both still and moving water in
fsoorlubelaech.extArlaicqtsu;ottsheofcaerqbuoahlyddrraytewpeeigahktfrwoemrethbeio4a%ssaNyaedOHforexatlrlacwtat(esre-e response to the dissolved cue (Tamburri etai, 1992; Turner
Fig. 4) was dissolved in water prior to the bioassay. Percentages of ct ill., 1994; Tamburri el <//., 1996). These results demon-
metamorphosis are means + SD (n = 3). strate that solublechemicals can increase settlementratesby
influencing the behavior of larvae still suspended in the
watercolumn (Turner etal, 1994). The relative importance
lecular weight are all active in promoting larval settlement, of the secretion of the Vaucheria-denved cue to the settle-
including those associated with structural elements of the ment of Alderia modesta larvae in the field will require
cell wall (Fig. 5). further study. However, the rapidity of the conditioning
process suggests that absorbent mats of V. longicaulis may
Chemical and enzymatic treatment ofboiled Vaucheria become saturated with naturally conditioned water during
extract (BVE) high tides, which might induce settlement in larvae that
To determine whether proteinacious components were enter trapped water parcels.
required for bioactivity, BVE was treated with a high con- notExotnrlayctdienmgonVsaturcahteerdiathaltontghiecacuhleimsicwailthcubeoiilsinsgtabwlaetetro
centration (2 mg/ml) of proteinase K. Proteinase treatment
BVE prolonged periods of boiling, but that a highly concen-
dwiadsnaoltsodesctraebalseetothmeiblidoaacctiidvihtyydrooflysis u(sFiign.g60)..1BMioaTctFiAvitayt trated solution can be prepared in this manner. The di-
100C. In contrast, treatment with sodium periodate. a minishing bioactivity of sequential extracts indicates that
chemical agent which cleaves sugar residues, significantly
decreased the bioactivity of BVE (Fig. 6).
Discussion
The present study demonstrates that larvae of the asco-
a
glossan Alderia modestu metamorphose in response to both
secreted and surface-associated forms ofthe settlement cue
derived from the adult host alga, Vaucheria longicaulis.
Seawatei conditioned by the presence of V. longicaulis was
as active as the algae itself at inducing larval metamorpho-
sis, suggesting the algae secretes one form of the cue into
the surrounding seawater. Water-soluble cues derived from Figure 6. Effects of proteinase K. mild acid hydrolysis, and sodium
the adult prey organism induce larval settlement for other periodateon bioactivityofboiled Vaucheriuextract(BVE). Percentagesof
specialist predators. Larvae of the aeolid nudibranch Phes- larval metamorphosisaregivenasmeans + SD(n = 3)foreachtreatment
and the correspondingcontrol. Percentages foreach treatment and control
lilln \ihoi>ac metamorphosed in response to water condi- werearcsinetransformedandcomparedwithatwo-tailedimpairedt-test(*
tioned with the hard coral Porites compressa (Hadfield, = significantatP<0.05level);thetvaluesobtainedwere2.08(proteinase
1977; Hadfield and Scheuer, 1985). Larvae of the dorid K), 1.20 (mild acid hydrolysis), and -4.46 (sodium periodate).
CARBOHYDRATE SETTLEMENT CUES 101
there is a limited amount of the waterborne cue that can charide units of polysaccharides, oxidizing consecutive
be extracted with boiling water. However, after repeated hydroxyl groups to aldehydes and cleaving sugar residues
extractions the residual Vaucheria tissue retained signif- having three consecutive hydroxyl groups to produce
icant activity, indicating that a non-extractable form of formic acid (Hassid and Abraham, 1957). Taken together,
the settlement cue remains associated with the algal cell the data strongly suggest that the larvae of Alderia ino-
wall. This is the first direct demonstration that the same desta metamorphose in response to a structural feature of
substrate produces both secreted and surface-associated the polysaccharides produced by Vaucheria longicaulis.
forms of a larval settlement cue, each of which is suffi- This would account for the bioactivity of molecules of
cient to induce metamorphosis. differing molecular weight, since small oligosaccharides
Polysaccharide chemists routinely employ basic solvents can contain the same distinctive glycosidic linkages as
such as NaOH to extract material that remains associated are found in the full-length polymer. Consistent with this
with plant cell walls following hot water extraction (Cleare hypothesis, the activity of Vaucheria extract was not
and Percival, 1972). The water-insoluble material extracted diminished by a mild acid hydrolysis using 0.1 M TFA;
with 4% NaOH contained high molecular weigMht carbohy- the same conditions have been used to fragment matrix
drates that, when partially resolubilized in 7 urea and polysaccharides of green algae into smaller oligosaccha-
tmheenntdiaasslayyz.edTahgeainbsitoawcattievri,tyweorfethaicstifvreaicntitohneilsarcvoanlssiestttelnet- rides representative of the repeating unit (Lahaye and
with the finding that Vaucheria tissue exhaustively ex- Ray, 1996).
tracted with boiling water can still induce metamorphosis; Recognition of carbohydrates by larval lectins has been
together, these experimentsdefine asurface-associatedclass implicated in settlement induction forseveral taxonomically
of molecules which differ in their physical properties (size. diverse marine invertebrates (Kirchman et al., 1982; Maki
solubility) from the water-soluble cue molecules, but share and Mitchell, 1985; Bahamondes-Rojas and Dherbomez,
the same bioactivity. An insoluble inducer associated with 1990; Bonar et al., 1990; Morse and Morse, 1991). Meta-
cell wall polysaccharides of the crustose red alga Hydroli- morphosis ofbarnacle larvae in response to glycoproteins is
thon boergesenii triggered metamorphosis in the coralAga- abolished when the oligosaccharide chains of the proteins
ricia humilis (Morse et al, 1988; Morse and Morse, 1991 ). are boundby lectins and thus rendered inaccessible to larval
Larvae of the echinoid Stronglyocentrotus droebachiensis receptors (Matsumura et al., 1998). The present study
metamorphosed in response to live tissue or a homogenate strongly indicates a carbohydrate is the settlement cue for
of several species of coralline red algae, but the algae did Alderia modesta, but definitive proofwill require the isola-
not release soluble inducers into seawater (Pearce and tion of a pure oligosaccharide that induces metamorphosis.
Scheibling. 1990). Preliminary results indicate that inductive fragments are
The active material in the water-soluble extracts was anionic and contain uncommon sugar residues including
further divided into discrete molecular weight classes by glucuronic and galacturonic acid, rhamnose, and xylose,
ethanol precipitation. Chromatography revealed that the which are not recognized by most available enzymes and
carbohydrates in the BVE precipitate were exclusively of lectins (Krug, 1998a). A direct chemical analysis of the
high molecular weight, while those in the supernatant structural features ofthe polysaccharides of Vaucheria lon-
were all of low molecular weight; in both cases, the gicaulis and their bioactivity is currently underway. How-
bioactivity co-eluted with the major carbohydrate peak. ever, bioactivity is clearly associated with algal polysaccha-
Studies of larval settlement inducers for other opistho- rides, both soluble and insoluble, making A. modesta an
branchs have used ultrafiltration to show that the bioac-
tive molecules are less than 1,000 Da in size (Hadfield ideal experimental organism for dissecting the roles of
waterborne versus surface-associated cues in the larval set-
and Pennington, 1990; Gibson and Chia, 1994; Lambert
et al., 1997). Distinct size-classes of water soluble set- tlement process.
tlement cue molecules have not been previously reported
from other study systems.
Acknowledgments
The bioactive settlement cue molecules co-eluted with
the carbohydrate peak in each extract, were stable to
boiling and mild acid or base treatment, and some were We thank Dr. K. Norgard-Sumnicht for experimental
firmly associated with the algal cell wall. These results assistance, and Drs. N. Holland, L. Levin, W. Fenical, C.
suggested that the molecules were either composed of, or Derby, and two anonymous reviewers for thoughtful criti-
tightly associated with, algal carbohydrates. Proteinase K cisms that greatly improved this manuscript. Access to the
treatment did not diminish the activity of algal extracts, Kendall-Frost Reserve was made possible by Isabelle Kay
but bioactivity was significantly reduced by treatment and the University of California Natural Reserve System.
with sodium periodate. Periodate reacts with monosac- P. J. K. was supported by an NSF Predoctoral Fellowship.
102 P. J. KRUG AND A. E. MANZI
Literature Cited Hassid, W. Z., and S. Abraham. 1957. Chemical procedures foranal-
ysis ofpolysaccharides. Methods in Enzymol. 3: 34-50.
Abbott, I. A.,and G.J. Hollenberg. 1976. MarineAlgaeofCalifornia. Hurlhut, C. J. 1991. Community recruitment: settlement and juvenile
Stanford University Press. Stanford. survival ofseven co-occurring species ofsessile marine invertebrates.
Bahamondes-Rojas,I.,andM.Dherbomez. 1990. Purificationpartielle Mar. Biol. 109: 507-515.
de substances glycoconjuguees capables d'induire la metamorphose Keough, M. J., and B. J. Downes. 1982. Recruitment ofmarine inver-
des larves competentes d'Eitbrancluts doriae (Trinchese. 1879), mol- tebrates:theroleofactivelarvalchoicesandearlymortality. Oecologia
lusque nudibranche. J. E\p. Mar. Biol. Ecol. 144: 17-27. 54: 348-352.
Bonar,D. B.,S. L. Coon,M. Walch,R.M. Weiner,andW. Fitt. 1990. Kirchman. D., S. Graham, D. Reish, and R. Mitchell. 1982. Lectins
Control of oyster settlement and metamorphosis by endogenous and may mediate in the settlement and metamorphosis ofJanita (Dexio-
exogenous chemical cues. Bull. Mar. Sci. 46: 484-498. spira) brasiliensis (Grube). Mar. Biol. Lett. 3: 131-142.
Burke. R. D. 1983. The induction ofmetamorphosis ofmarine inverte- Krug, P. J. 1998a. Chemical and larval ecology ofopisthobranch mol-
brate larvae: stimulus and response. Can. J. Zoo/. 61: 1701-1719. luscs: variable development modes and settlement requirements for
Burke. R. D. 1984. Pheromonal controlofmetamorphosisin the Pacific larvaeofAlderiamodesta. Ph.D.dissertation. UniversityofCalifornia,
sand dollar. Dendrasterexcentricus. Science 225: 442-443. San Diego. 227 pp.
Bulman. C. A. 1987. Larval settlement of soft-sediment invertebrates: Krug, P. J. 1998b. Poecilogony in an estuarine opisthobranch: plank-
the spatial scales of pattern explained by habitat selection and the totrophy, lecithotrophy, and mixed clutches in a population of the
emergingroleofhydrodynamicprocesses. Oceanagr.Mar. Biol.Ainni. ascoglossanAlderia modesta. Mar. Biol. 132: 483-494.
Rev. 25: 113-165. Lahaye, M., and B. Ray. 1996. Cell-wall polysaccharides from the
NMR
Butman. C. A., and J. P. Grassle. 1992. Active habitat selection by marine alga Ulva "rigida" (Ulvales, Chlorophyta) analysis of
Capilella sp. I larvae: two-choice experiments in still waterand flume ulvan oligosacchandes. Carbohydr. Res. 283: 161-173.
flows. / Mar. Res. 50: 699-715. Lambert,W.J.,andC.D.Todd. 1994. Evidenceforawater-bornecue
Butman,C.A.,J. P.Grassle,and C.M.Webb. 1988. Substratechoice inducing metamorphosis in the dorid nudibranch mollusc Adalaria
madebymarinelarvaesettlinginstillwaterandinaflumeflow.Nature pntxima (Gastropoda: Nudibranchia). Mar. Biol. 120: 265-271.
333: 771-773. Lambert, W. J., C. D. Todd, and J. D. Hardege. 1997. Partial char-
Chia, F.-S., and R. Koss. 1988. Induction of settlement and metamor- acterization and biological activity of a metamorphic inducer of the
phosisofthe veligerlarvae ofthe nudibranch Onchidorisbilamellata. dorid nudibranch Adalaria proximo (Gastropoda: Nudibranchia). In-
Int. J. Invrrtebr. Reprod. Devel. 14: 53-70. vertebr. Biol. 116: 71-81.
Cleare, M., and E. Percival. 1972. Carbohydrates of the fresh water LeTourneux, F., and E. Bourget. 1988. Importance of physical and
algaTribonemaaec/uale. I.Lowmolecularweightandpolysacchandes. biological settlement cues used atdifferent spatial scalesby the larvae
Br Phycol. J. 1: 185-193. ofSemibalanus balanoides. Mar. Biol. 97: 57-66.
Day,R.W.,andG.P.Quinn. 1989. Comparisonsoftreatmentsafteran Levin, L. A., and T. S. Bridges. 1995. Pattern and diversity in repro-
DuBoaniasl,ysMi.s,ofK.vaAr.iaGniclelesin,eJ.coKl.ogHy.amEiclotlo.nM,oPn.ogAr..R5e9b:er4s3,3-a4n6d3.F. Smith. dMuccEtdiwonaradn,dedde.veClRopCmePnrte.ssP,p.Oxf1or1d7.inMarineInvertebrateLarvae. L.
1956. Colorimetric method for the determination of sugars and re- Makise,ttJl.emeS.n,taannddmeRt.amMoirtpchhoeslli.so1f9m85a.rineInivnovlervteembernatteolfarvlaeec.tiBnusll.inMatrh.e
lated substances. Anal. C/iem. 28: 350-356.
Sci. 37: 675-683.
Eckman,J. E. 1983. Hydrodynamic processes affecting benthic recruit- Maki,J.S.,D.Rittschof,M.-O.Samuelsson,U.Szewzyk,A.B.Yule,S.
Eckmmeannt.,JL.imEn.ol1.99O0.ceanAogmro.de2l8:of24p1as-s2i5v7e.settlementby planktoniclarvae bKajcetlelreibaeragn,dJ.thDei.rCeoxsotploolwy,maernsdoRn. Mtihtechaetltla.ch1m9e9n0t.ofEfbfaercntacolfemcayrpinnse
Gibsoonnt,obGo.ttDo.m,saonfddiFf.f-eSr.ingChrioau.gh1n9e9s4s.. LiAmnomle.taOmcoerapnhoigcr.in35d:uce8r87i-n90t1h.e Matlsaurvmauer.aB,ullK..,MaSr..MoScrii.,46M:.4N9a9g-a5n1o,1. and N. Fusetani. 1998. Lentil
opisthobranch Hwninaeacallidegcnita: partial purificationandbiolog- lectininhibitsadultextract-inducedsettlementofthebarnacle.Balanus
ical activity. Biol. Bull. 187: 133-142 amphitrite. J. Exp. Zoo/. 280: 213-219.
Grahame, J., and G. M. Branche. 1985. Reproductive patterns of Morse,A.N.C.,C.A. Froyd,and D.E. Morse. 1984. Moleculesfrom
marine invertebrates. Oceanogr. Mar. Biol. Anna. Rev. 23: 373-398. cyanobacteriaandredalgaethatinduce settlementand metamorphosis
Grosberg, R. 1982. Intertidal zonation of barnacles: the influence of in the mollusc Haliotis rufescens. Mar. Biol. 81: 293-298.
planktonic zonation of larvae on the vertical distribution of adults. Morse,D. E.,andA.N.C.Morse. 1991. Enzymaticcharacterizationof
Ecology 63: 894-899. the morphogen recognized by Agaricia humilis (Scleractinian coral)
Hadfield, M. G. 1977. Chemical interactions in larval settling of a larvae. Biol. Bull. 181: 104-122.
marinegastropod.Pp.403-413inMarineNaturalProductsChemistry. Morse, D. E., N. Hooker, A. N. C. Morse, and R. Jensen. 1988.
D. J. Faulkner and W. H. Fenical, eds. Plenum. New York. Controloflarvalmetamorphosisandrecruitmentinsympatricagariciid
Hadfield, M. G., and J. T. Pennington. 1990. Nature ofthe metamor- corals. / Exp. Mar. Biol. Ecol. 116: 193-217.
phic signal and its internal transduction in larvae of the nudibranch Pawlik, J. R. 1992. Chemical ecology of the settlement of benthic
Pheslilla sibogac. Bull. Mar. Sci. 46: 455-464. marine invertebrates. Oceanogr. Mar. Biol. Annu. Rev. 30: 273-335.
Hadfield, M. G., and D. Scheuer. 1985. Evidence for a soluble meta- Pawlik,J.R.,C.A.Butman,andV.R.Starczak. 1991. Hydrodynamic
morphic inducerin Phcxlilla: ecological,chemicalandbiologicaldata. facilitation of gregarious settlement of a reef-building tube worm.
Bull. Mar. Sci. 37: 556-566. Science 251: 421-424.
Hartog, C. Den. 1959. Distribution and ecology of the slugs Alderia Pearce,C.M.,and R. E. Scheibling. 1990. Inductionofsettlementand
modesla and Limapontiu dcpressa in the Netherlands. Beaufortia 7: metamorphosis inthe sanddollarEchinarachniusparma: evidence for
15-36. an adult-associated factor. Mar. Biol. 107: 363-369.
Hartog,C. Den.,and C. Swennen. 1952. On the occurrence ofAlderia Raimondi,P.T. 1988. Settlementcuesanddeterminationofthevertical
modesta(Loven)andLimapontiadepressaA.& H.onthesaltmarshes limit ofan intertidal barnacle. Ecology69: 400-407.
ofthe Dutch Wadden Sea. Beaufortia 2: 1-3. Roberts, D., D. Rittschof, E. Holm,and A. R. Schmidt. 1991. Factors
CARBOHYDRATE SETTLEMENT CUES 103
influencing initial larval settlement: temporal, spatial, andsurface mo- Todd, C. D., M. G. Bentley, and J. N. Havenhand. 1991. Larval
lecularcomponents. J. Exp. Mar. Biol. Ecol. 150: 203-221 metamorphosis ofthe opisthobranch molluscAdalariaproximo (Gas-
Rodriguez,S.,F.PatricioOjeda,andN.Inestrosa. 1993. Settlementof tropoda: Nudibranchia): the effects ofcholine and elevated potassium
benthic marine invertebrates. Mar. Ecol. Prog. Ser. 97: 193-207. ion concentration. / Mar. Biol. Assoc. U. K. 71: 53-72.
Roughgarden, J., S. Gaines, and H. Possingham. 1988. Recruitment Trowbridge, C. 1993. Local and regional abundance patterns of the
dynamics in complex life cycles. Science 241: 1460-1466. ascoglossan opisthobranch Alderia modesta (Loven, 1844) in the
Tamburri, M. N., R. K. Zimmer-Faust, and M. L. Tamplin. 1992. Northeastern Pacific. Veliger36: 303-310.
Natural sources and properties of chemical mducers mediating Turner, E. J., R. K. Zimmer-Faust, M. A. Palmer, and M. Lucken-
settlement of oyster larvae: a re-examination. Biol. Bull. 183: 327- back. 1994. Settlement ofoyster (Crassostrea virginica) larvae: ef-
338. fectsofwaterflow and a water-soluble chemical cue. Limnol. Ocean-
Tamburri, M. N., C. M. Finelli, D. S. Wethey, and R. K. Zimmer- ogr. 39: 1579-1593.
Faust. 1996. Chemical induction of larval settlement behavior in Underwood, A. J.,and P. G. Fairweather. 1989. Supply-side ecology
flow. Biol. Bull. 191: 367-373. and benthic marine assemblages. TREE4: 16-20.