Distribution and Characterization of Ion Transporting and Respiratory Filaments in the Gills of Procambarus clarkii

13 Pages·1991·5.2 MB·English

by  J S Dickson

Reference: Bin! Bull 180: 154-166. (February. ll91)
Distribution and Characterization of Ion Transporting
and Respiratory Filaments in the Gills
of Procambarus clarkii
JOHN DICKSON, RICHARD M. DILLAMAN,
S.
ROBERT D. ROER, AND DAVID B. ROYE
Centerfor Marine Science Research, University ofNorth Carolina at Wilmington,
7205 n'rightsvi/le Ave., Wilmington, North Carolina 28403
Abstract. Individual gill filaments of the freshwater volved in the separate but interrelated functions of gas
crayfish Procambarus clarkiiweredetermined to beeither exchange, acid/base balance, and ion regulation. Those
predominantly respiratoryortransporting. Silverstaining substances exchanged may include oxygen, electrolytes,
revealed that the filaments within the central bed ofthe carbon dioxide, and ammonia. To produce a large surface
gills formed silver deposits whereas filaments at the mar- area forsuch exchange, gillsareoften branchedalongone
ginsand theentiresixth pleurobranch formed nodeposits. or more axes. In the crabs(Crustacea, Decapoda) lamellar
Designation ofthe silver staining gills as predominantly gills are present (Copeland, 1963, 1968; Copeland and
transporting and unstained filaments as predominantly Fitzjarrell. 1968;TaylorandGreenaway, 1979; Finol and
respiratory was substantiated by ultrastructural analyses Croghan, 1983) and consist ofa parallel afferent and ef-
and measurements ofATPase and transepithelial poten- ferent blood vessel between which are broad flattened
tials. Presumptive transporting filaments had an epithe- sheets. In branchipod crustaceans, such Artemia salina
lium subjacent to the cuticle that was relatively thick and andDaphnia nuigna(Copeland, 1967 and Kikuchi, 1983),
dominated byabundant mitochondria. Lacunaewerede- amphipods and isopods(Milneand Ellis, 1973; Bubel and
lineated by pillar structures and served ascollateral path- Jones, 1974)gillsare flat, oval, sac-likeextensions oftho-
ways for the movement ofblood from the afferent to ef- racic appendages or modified pleopods. In crayfish and
ferent blood channels, which were separated by a thin shrimp the gills are filamentous (Morse el a/.. 1970; Bie-
septum. Presumptive respiratory filaments had an ex- lawski, 1971; Fisher, 1972; Burggren el ai. 1974; Foster
tremely thin epithelium with few organelles, but a rela- and Howse, 1978). Incrayfishthegillsaretrichobranchiate
tively thick septum. Present in both types of filaments and have a central stalk from which hundreds of tiny
werenervesand podocytes. ThevaluesforNa, K-ATPase finger-like filaments arise. The podobranchia (outermost
weresignificantly higherin thetransportingfilamentsthan gills)areattachedtothecoxaeoftheappendages. A mem-
in those designated as respiratory. The measurement of branous lamina, devoid of filaments, extends from the
transepithelial potentialsshowed both filamentstobecat- innerside ofthe podobranch and liesagainst thethoracic
ion selective with the respiratory filaments slightly more body wall; its distal tip is flattened into a plate bearing a
positiveand thetransporting filaments slightly more neg- few filaments. The arthrobranchia, which are attached to
ative than the diffusion potential for Na. the articular membrane between the body wall and the
appendages, and the pleurobranchia, which are attached
Introduction to the pleural wall ofeach somite, lie beneath the podob-
ranchia and have no distal plate structure (Huxley, 1986;
Gills are a site for exchange between the blood ofthe Burggren et ai, 1974).
organism and the external medium and, as such, are in- Despite variations in gill anatomy, the ultrastructure
ofthe gill epithelium in those species that are capable of
Received 28 June 1990;accepted 9 November 1990. osmoregulation is similar, and can be placed into one of
154
GILL FILAMENTS IN P Cl.ARkll 155
two categories. Both categories consist ofa single layered elal. ( 1982). on theotherhand, observedonly respiratory
epithelium that separates blood spaces from the uncal- epithelial tissues in the filamentsand ion transportingep-
cified cuticle. One type of epithelium is relatively thick ithelial tissues in the podobranch lamina ofAstacuspal-
and has the characteristics ofan ion transporting epithe- lipes and therefore assigned the appropriate function to
lium (seeCioffi. 1984, forareviewofarthropod ion trans- the two structures. The only ultrastructural observations
porting epithelia). Briefly, it consists ofcells with highly of filaments have been of transporting epithelia (Morse
folded apical and basal processes, aprominent basal lam- et al.. 1970; Burggren et al.. 1974; Bielawski, 1971; Fisher,
ina and cytoplasm containing abundant mitochondria as 1972).
well as golgi apparati, smooth and rough endoplasmic This investigation was therefore directed at determin-
reticulum and glycogen. The second type ofepithelium ing within a single filament, within a single gill, and
consists ofvery thin squamous cells subjacent to the cu- among gills the distribution and frequency ofthe two
ticle. The cells are only one tenth the thickness of the types ofepithelia ofthe freshwater crayfish Procambarus
previouslydescribedtype,andthecytoplasmcontainsvery chirkii. The methods employed in this study were silver
few organelles (Copeland and Fitzjarrell, 1968). This ul- staining (Maluf, 1940), an indicator of ion transporting
trastructure would appear to favor diffusion ofgases be- tissue; transmission electron microscopy (TEM), which
tween the external and internal media and consequently revealsthe strikingdifference between the morphology of
this epithelium has been categorized as being respiratory the two types ofepithelia; analysis for Na, K-ATPase, an
in function. effector ofion movement; and measurement oftransep-
Studies on a variety ofspecies ofdecapod crustaceans ithelial potentials in individual filaments, which reflects
have demonstrated the existence ofa partitioning ofion possible differences in transport or permeability of the
regulatory and respiratory functions amongdifferent gills combined epithelial and cuticular layers.
or within different regions of gills. Both morphological
andbiochemical(Na, K-ATPasedistribution)studieshave Materials and Methods
provided evidence that the anterior gills are dedicated to
gas exchange while the posterior gills are involved in ion The crayfish (Procambarus clarkii), obtained from
regulation in theeuryhaline brachyurans Eriocheirxinen- Carolina Biological Supply, were kept in aquaria ofarti-
sliisda(y,Pe1q9u85e)u,xUacnad/Gni/lmliesv,(W19a7n8s,on19e8t1a)l,..U\c9a84p)u,gCanlalxin(eHcotle-s bfiectiawlepenon2d0wCataenrd(R2o5eCr,aannddSfheedlt2o-n3, t1i9m8e2s),amwaeienkt.aOinnleyd
sapidus (Copeland and Fitzjarrell, 1968; Neufeld et al., active crayfish in the intermolt stage, as determined by
1980; Towle, 1984), and Carcinus maenas (Mantel and thecriteria ofDrach and Tchernigovtzeff(1967) and Ste-
Landesman, 1977). Although we and other investigators venson (1972), were used in this study.
realize that gill tissues possessing transport epithelia are
also engaged in gas exchange, and that ions move across Silverstaining
respiratory surfaces, we still use. for simplicity, the pre- Crayfish subjected to silverstaining were first rinsed in
dHeonmcienfaonrtthf,uwncetiwiolnltroefdeerstiogntahtoesethgialtloffiltahmeeennttsitrheaftiploasmesnets.s datei2on5iCzedforwa3t0ermiann.dTphleacveodliunmea 0of.0t5he%sAolguNtiOon3 swoalsutsiuofn-
a typical transporting epithelium as "transport" typesand ficienttocompletelysubmergethecrayfish. Crayfish were
those gill filaments with a squamous epithelium as "re- then rinsed with deionized water. The branchiostegites
spiratory" types. were removed to expose the gills, and the crayfish were
Crayfish, like theeuryhaline brachyurans in dilute me- placed in Kodak Microdol-X developer (diluted 3:1 to
dia, exist in a hypoosmotic medium, and must maintain lower the osmolarity to 630 mOsm) for one hour with
their ion balance by actively transporting electrolytes frequent agitation. Afterthegillswerethoroughlywashed
across the gills (Maluf, 1940; Bryan, 1959; Shaw, 1960a. and placed in distilled water, one or two drops ofNH4S
b; Ehrenfeld, 1974). Unlike the euryhaline brachyurans were added to intensify the stain.
studied, however, mostcrayfish arestenohaline freshwater
crustaceansandpossessfilamentous, ratherthan lamellate,
gills. The partitioning ofrespiratory and ion transporting fPorrepealrecattrioonnmoijcgriollsfciolpayment tissues
function in the gills ofcrayfish is still in question. While
Wheatly and Henry (1987) have used biochemical anal- Silvertreated filamentswere fixed in 3% glutaraldehyde
M
ysesofpooled gill setstoshow thatthedistribution ofNa, in0.1 cacodylate buffer(pH 7.2)containing 5% sucrose
K-ATPase was similar among gills of Pacifastacus len- (osmolarity = 640 mOsm) at 25 C for at least 1 h. To
iusculus, Morse et al. (1970) used silver staining and re- facilitate penetration ofthe varioussolutions, thetipswere
portedapartitioningofrespiratoryandtransportfunctions cut off"the filaments and the tissue wasgently agitated at
among the gill filaments ofthe same species. Dunel-Erb each step of the preparation. The filaments were then
156 J. S. DICKSON ET AL
rinsed iMn buffer and postfixed with 2% osmium tetroxide Transepithetialpotential measurements
in 0.1 cacodylate buffer (pH 7.2) plus 5% sucrose at
od2xe5ihdyCed,rfaoatrned2dhew.miAtbfhetdeardgeardnaodtienhdeSsrperruiirenrsseolfwoiwetthvhiabsnucofolfseiart,nydteimpssrbuoeepdsydwlieenrngee RKiCnC1grae=yrfis1so.3hluwmteAiro/ne;s(CfuaibnCamlle:rcog=necd3e.ni4tnrm'aXAit/si;torneMsn:ggCtNhlaV2Cal=n=0H.a35r1rm.e3Av/em)lAd/'i;sn
medium a finger bowl, and restrained by rubberbandsattached to
(Spurr, 1969).
dehFyidleamiennt0s.1n5otMsiclavceord-tyrleaatteedbuwfefrere,fpixHed7.i4n,2w%itghlu5tamra.lM- ausneodtctohepdropvleixdieglsausfsfipcliaetnet.eDlielcuttroeldytReisngfeorr'selseoclturtoidoencwoans-
0C.a5C%l2OfsorO41 whi,thri0ns.e8d%iKn4bFueff(eCrN,)a6ndinp0o.s1t-Mfixecdaciondyflraetseh dteurcntailonm.ePdoiteunmtiwalesremeuanrseulrieabdlew.itThhpeobnrdanwcahtiearlarsegtihoenexo-f
buffer, pH 7.4, with 5 mA/CaC!2 at 25C for 1 h. After the carapace was cut away from one side ofthe animal
rinin0s.i1n5g%agtaainnniinccaaccidodiynla0.te15buMffecra,cotdhyeltaitsesubeusffweerr,eppHla7c.e4d, ctoouelxdpoesaesitlhyebgeillis.mpInaltehidsbpyosigtliaosns,miincdrioveildeucatlrgoidllesfiulsamienngtsa
with 5 mA/ CaCl2 for 5 min. Following a short rinse in micromanipulator.
bufferand distilled water, the tissueswere stained en bloc Microelectrodes were drawn from glasscapillary tubing
with 2% aqueous uranyl acetate for 2 h. After a distilled and filled with 3 A/ KC1. A Ag-AgCl reference electrode
water rinse, tissues were dehydrated through a graded se- was immersed in the bathing medium. Both electrodes
ries ofacetone and embedded in Spurr. were connected by shielded cables to a WPI model 701
Cross-sections were cut in the basal, medial, and distal microprobe system and a Tectronix 5111 storage oscil-
regions ofthe filaments. Sections were post-stained with loscope.
4% uranyl acetate in 50% ethanol and Reynolds lead ci- Only filaments ofthe podobranchs were impaled and
trate (Reynolds, 1963), and were viewed with a Zeiss EM the assignment ofputative transport and respiratory fil-
9S transmission electron microscope operated at 60 kV. aments was based upon the results ofthe silver staining
experiments. Transepithelial potentials (TEP's) and the
position ofthe electrode tip within the filament were re-
Na, K-ATPase assay offilaments corded each time a gill was impaled. Sodium concentra-
Na, K-ATPase assays were done according to the tions ofthe medium and crayfish hemolymph were mea-
methoMd ofHoriuchi (1977). Afterwashingexcised gills in sured by flame photometry (Turner model 510).
0.25 sucrose buffered to pH 7.5 with 100 mAl Tris-
HC1, predicted transportingor respiratory filamentswere Results
re4moCved from thegills, pooledbytype, and homogenized Silver staining
at in 2volumesofthebuffered sucrosesolutionduring
35 passes in aglasstissue homogenizer. The homogenate In six crayfish subjected to silver staining all showed
was centrifuged at 3000 X g for 15 min and the super- some staining in all gills except pleurobranch 6. Staining
natant was assayed for activity. occurred only in the filaments. That is, no staining was
The reaction mixture consisted of 60 mA/ NaCl. 20 observed in the central gill stalk, in the lamina, in the
mA/ KC1, 2 mA/ MgCl2 and 30 mA/ Tris-HCl, pH 7.5, distal plate of the podobranchia, nor at the base of the
in0.8 ml. This reaction mixturewaspreincubated at 32C gills. Figure 1 is representative of the filament staining
before 0.1 ml ofthe enzyme solution and 0.1 ml ofthe patterns. The podobranchs had the largest proportion of
3.0 mA/ Tris-ATP were added. The enzymatic reaction stained filaments, whereasthe pleurobranchs had the least
was allowed to proceed for 45 min, being terminated by foreach setofgills. Allgillsshowedthesamebasicpattern,
theaddition ofcold 3% trichloroacetic acid. The amount a population of stained filaments in the central area of
ofinorganic phosphate, hydrolyzed from ATP, present in the filament bed surrounded by lateral rows ofunstained
the supernatant was measured according to Wheeler filaments, with the exception ofthe pleurobranch ofthe
(1975). Mg-ATPase activitywasdetermined asabovewith sixth gill set, which showed no silver staining. Silver
thesubstitution of10 mA/ouabain forKC1 in thereaction stained filaments had precipitate distributed evenly over
mixture to inhibit Na, K-ATPase. Na, K-ATPaseactivity the length ofthe filament.
wascalculated asthedifference between thetotal ATPase Transmission electron microscopy ofsilver stained fil-
and Mg-ATPase values. The amount ofprotein was de- aments showed that most of the precipitate was within
termined according to Peterson's (1977) modified Lowry the gill cuticle, although lesser amounts were observed
protein assay using a Bausch and Lomb Spectronic 710 outside or beneath the cuticle (Fig. 2). Precipitate within
spectrophotometer and bovine serum albumin as a stan- thecuticle was localized almost exclusively inalayercor-
dard. The specific activitieswere recorded as micromoles responding to the exocuticle, rarely being found in the
ofinorganicphosphatepermilligram ofprotein perhour. epicuticle or endocuticle. While the treatment employed
GILL FILAMENTS IN P. CLIRKII 157
Podobranch Arthrobranch Pleurobranch
Gill Set
Gill Set #2
GUI Set #3
Gill Set
Gill Set
Gill Set #6
Figure I. Light micrographsofsixsetsofgillsfrom therightsideofacrayfishaftersilverstaining. Each
gill isoriented with the basetothe right. Dark stainingindicatessilverdeposition.
in the silverstainingdisrupted the soft tissues, itwas pos- upon the distribution of the two types as indicated by
sible to note that the epithelium underlying the cuticle silver staining(Fig. 1)and proved to be accurate in all 20
was relatively thick and had numerous basal and apical ofthe filaments examined. No differences were observed
infoldings. Filamentscontaining no precipitate aftersilver among sections from proximal, medial, or distal regions
treatment (Fig. 3) had an epithelium subjacent to the cu- ofa filament. Diagrammatic representations ofcross sec-
ticle that was markedly thinner than that of the silver tionsthroughthetwotypesoffilamentsareseen in Figures
stained filaments and had few, if any, basal and apical 4 and 5. The transporting filaments, regardless ofthe gill
mm
infoldings. set, had a diameter of0.1 to 0.2 mm, while the re-
mm mm
spiratory filaments were 0.2 to 0.3 in diameter.
Ultrastructure ofthegillfilaments Both types offilaments contained an afferent and an ef-
ferent blood channel and lateral lacunae. The afferent
Choice offilaments for ultrastructural analysis ofpre- channel was on the side ofthe filament that faced toward
sumptiveion-transportingand respiratorytissuewasbased the gill stalk, and the efferent channel on the side that
158 J. S. DICKSON ET AL
Figure 2. TEM oftransportinggill filamentaftersilverstaining. Notecrystalswithinandon thecuticle
(arrows). Bar = 5 nm.
Figure3. TEM ofrespiratorygill filament aftersilverstaining. Bar = 1.0 pm.
facedaway from thestalk. A septum and thesurrounding The epithelial cells forming the pillar structures had a
connective tissue separated the two major blood spaces. highlyconvoluted basal nucleus(Fig. 7)and were usually
An epithelium, 80-90% ofwhich occurred in the lacunar rich in mitochondria, rough endoplasmic reticulum, and
bloodspace, linedthenoncalcified cuticleofthe filaments. golgi apparati (Fig. 8).
Epithelial cellsextendedacrossthelacunae formingpillar Thecuticleofthetransportinggill filamentswasofuni-
structures with their basal portion embedded in loose form thickness(approximately 1.8j/m)aroundtheirentire
connective tissue. Also depicted is the marked difference circumference. An outerthin epicuticle, thethin lamellae
in epithelial thickness within and between the filament of the exocuticle, and the thicker lamellae ofthe endo-
types. cuticlewereeasilydifferentiated in most sections(Figs. 6,
In the transporting filaments, the cuticle epithelium 9). Theapproximate thicknessesoftheexocuticleand the
bordering the afferent channel was the thickest (approx- endocuticle were 0.8 ^m and 0.6 ^m, respectively.
imately6 ^m)(Fig. 4). Numerous mitochondria, generally The septum traversed the central blood space,joining
oriented perpendicular to the cuticle, were observed and the loose connective tissue on either side ofthe filament,
weremostabundant in theapical portion ofthecells(Figs. thereby forming a partition between the afferent and ef-
6, 7, 8). Apical microvillar processes were approximately ferentbloodchannels(Figs. 4, 5, 10, 16). Alongtheefferent
80 nm wide and varied in length from 0.4 ^m to 1.2 pm channel the septum was relatively smooth, whereas the
(Fig. 8). Electron-dense material was observed between side facing the afferent channel possessed processes that
the tips of the microvilli and the overlying cuticle (Fig. extended into the channel (Figs. 10, 16). In the trans-
9). Numerous infoldings and interdigitations occurred in porting filaments the septum was relatively thin (approx-
the basal region ofthe epithelium (Figs. 6, 7). The micro- imately 1.1 ^m) (Fig. 10) and became shorter as the fil-
tubules observed within the cytoplasm were oriented ament tapered in its distal portions. Numerous mem-
nearly perpendicularto thecuticleand were often closely braneswereobservedalongthelength oftheseptum, some
apposed to mitochondria (Fig. 8). Occasional golgi ap- of which constituted the cell membranes of loose con-
parati were observed in the epithelium lining the outer nective tissue cells on either side ofthe filament. A thin
margin oftheafferentcanal (Fig. 6), aswell asathin basal basal lamina wasobserved on the efferent side ofthe sep-
lamina (Figs. 6, 9). Adheringjunctions occurred apically tum, and a thicker fibrous basement membrane covered
on lateral cell membranes, below which were septate the afferent septal process (Figs. 10, 16).
junctions. Gapjunctionswereobserved belowtheseptate The epithelium of the respiratory filament varied in
junctions (Fig. 6). thickness, ranging from 0.7 ^m opposite the septum to
Theportion oftheepithelium lateral totheseptum and 3.0 nm in regions adjacent to pillar structures (Figs. 5,
on the efferent side ofthe filament was thinner (approx- 1 1 ). A few scattered round or oval mitochondria were
imately 1.5 nm) than that portion bordering the afferent observed in the granular cytoplasm whereas thicker re-
canal, but also contained numerous mitochondria and gions of the epithelium adjacent to the pillar cells also
microvillar processes (Fig. 9). had some glycogen granules (Figs. 12. 13). The regions of
GILL FILAMENTS IN P CLARkll 159
to extensions ofthe cuticle that penetrated into the epi-
thelial cells(Fig. 13). The pillarcytoplasm contained some
mitochondria, rough endoplasmic reticulum, and golgi,
but noticeably fewerthan in the transportingfilamenttis-
sues (Fig. 12).
The cuticle ofthe respiratory filamentswasthickest on
the side ofthe afferent channel (approximately 2.0 ^m)
(Fig. 14) and thinnest on the efferent side ofthe filament
(approximately 0.5 fim) (Fig. 15). The attenuation ofthe
cuticle from the afferent to the efferent side wasobserved
lateral to the septum ofthe filament. It appeared as ifno
endocuticle had been deposited on theafferent side ofthe
filament, the side facing the carapace.
Theseptum ofthe respiratory filament wasthickerthan
thetransporting filament septum, possessingafferent sep-
tal projections, athick afferent basement membrane, and
a thin efferent basal lamina. The septum had a width of
about 5 ^m, with an abundance ofcell membrane inter-
digitations (Figs. 5, 16). A thick layer offibrous material
frequently lined the inside ofthe cell membrane on the
afferent side of the septum (Fig. 16). Thicker septa ob-
served in these filaments usually had loose connective
tissue extending across the septum on the afferent side
(Fig. 5).
Neural tissuewasoften observed in both filamenttypes
(Fig. 17). The small nerveswere located within one ofthe
small blood spaces ofthe connective tissue lateral to the
septum. Usually therewasone nerveperfilament, though
afew filamentswereobserved that contained twoorthree
nerves, each in a separate blood space. The nerve was
continuous throughout the length ofa filament. Nerves
ranged from 1.0^m to 2.7 jum in diameterandcontained
twotosix neurons. The neurons, which ranged from 0.16
m
fim to 1.5 M in diameter, contained numerous, evenly
spaced neurotubules. No cell bodies, neurosecretory
granules, or synapses were observed in any section.
Figures -4 and 5. Diagrammatic representations of cross sections Also occurring in both types of filaments were cells
tthheroaufgfehreanttrbalnospoodrtcihnagnnIeFilg.(A4b))a.nedffreersepnitrbatlooroyd(cFihga.nn5)elgil(lEbf)i,laempeintt.heNloitume resembling "podocytes" (Morse et ai, 1970). They were
(E), cuticle(c), septum (s). podocyte(P), hemocyte (h). pillarstructure observed neartheseptum, attachedtothelooseconnective
(p), and lacuna(1). tissue bordering the afferent channel (Figs. 4, 5, 18) and
were surrounded by a basal lamina. These cells possessed
an interdigitating cell membrane, or pedicels, which
the epithelium lateral to the septum and on the efferent formed extracellular spaces between the pedicels and the
side ofthe filament were much thinner, having a width cell body. Much ofthecytoplasm contained anabundance
as small as 0.08 ^m (Figs. 11, 14). The epithelium was ofsmoothendoplasmicreticulum. Coatedpitsandvesicles
lined by a thin basal lamina and had a sparse cytoplasm wereoften observed alongornearthecell membraneand
containing microtubules. Fibrous material was also seen numerous dense granules, varying in size and number,
between the basal lamina and the epithelium. No apical occurred within the cytoplasm along with golgi apparati
microvillarprocessesorbasal infoldingswere observed in consisting ofthin, curved cisternae (Fig. 19).
the respiratory filament epithelia.
The most prominent feature ofthe cells in the pillar ATPase determinations
regionswasthebundlesofmicrotubulesobserved intheir
cytoplasm (Figs. 12, 13). These bundles were generally As shown in Figure 20, the mean fortotal ATPase ac-
oriented perpendiculartothecuticle, apparentlyattached tivitywashigherinthetransportingthan in the respiratory
160 J. S. DICKSON ET AL
Figures6-9. TEMot'lransportingfilaments. Figure6. Sectionshowingthecuticlewithitsthreelayers,
the epicuticle (arrow), the exocuticle (Ex) and the endocuticle (En). Also note a Golgi apparatus (g). the
basal lamina(b).andajunctionalcomplex(arrowhead). Bar = 1.0pm. Figure7. Sectionshowingapillar
structure(p) separatingtwo lacunarspaces(1) from an efferent blood channel (Eb). Note heterochromatic
nucleus(n).Bar= 5.0^m. Figure8. Sectionofepithelium undercuticleshowingmicrotubules(arrowheads)
and rough endoplasmic reticulum (r) in thecytoplasm. Also note the apical microvillarstructures(arrow)
immediatelybelowthecuticle. Bar = 1.0^m. Figure9. Section ofepithelium betweencuticleand lacuna
showingthe electron dense material at the tip ofthe microvillar processes(arrowhead) and the thin basal
lamina(b). Bar = 1.0pm.
GILL FILAMENTS IN P CLIRK1I 161
Ab
Figures 10-16. TEM oftransporting(10)and respiratory (11-16) filaments. Figure 10. Section ofa
septum inatransportingfilamentthat separatestheafferent(Ah) from theefferent(Eh)hloodchannel. Bar
= 2.0^11. Figure 11. Sectionofarespiratoryfilamentshowingthethinepithelium underlyingthecuticle.
Also note the pillar structure (p) and a hemocyte (h). Bar = 5.0 ^m. Figure 12. Pillar structure ofa
respiratoryfilamentshowingthebundlesofmicrotubulesrunningperpendiculartothecuticle(arrowheads).
Bar = 1.0 nim. Figure 13. Section showing relationship ofmicrotubules (m) to extensions ofthe cuticle
(arrowheads).Bar=0.2/jm.Figure14. Cuticleonsideofafferentchannel.Bar= I.O/im.Figure15. Cuticle
on side ofefferent channel. Bar = 1.0 ^m. Figure 16. Septum from respiratory filament separating the
afferentbloodchannel(Ah)from theefferentbloodchannel(Eb). Notethethinbasal laminaontheefferent
side(arrowhead)ascompared with thethick layeron theafferent side(arrow). Bar = 1.0/im.
162 J. S. DICKSON AT AL
Figures 17-19. TEM offilaments. Figure 17. Cross section ofnerve showing individual neurons(n)
havingevenlyspacedneurotuhules(arrowheads). Bar =0.5/jm.Figure 18. Podocyte(P)inlooseconnective
tissue (Ct). Note electron-dense inclusions (d). Bar = 5.0 ^m. Figure 19. Podocyte cytoplasm showing
pedicels(arrowheads)golgi (g)and basal lamina(arrow). Bar = 1.0^m.
filaments; however,thedifferencebetweenthemeanswas ament impaled and ofthe podobranch upon which the
not significant. Mean values for Mg-ATPase were very filament was located. The TEP's of filaments that were
similarbetween respiratoryandtransportingfilamentsand presumed toberespiratory(based upon thesilverstaining
were likewise not significantly different. The means for results) ranged from -6 to -18 mV (-1 1.9 3.2 mV,
Na, K-ATPase, in contrast, weresignificantly different (P mean S.D., n = 19). The TEP's measured in presump-
< 0.001, Mann-Whitney U test), with the transporting tive transport filaments ranged from -21 to -36 mV
filaments having more than five times as much mean ac- (-28.1 4.7 mV, mean S.D.; n = 21). There was no
tivity as the respiratory filaments. overlap in the frequency distributions of the ranges of
TEP's(Fig. 21)and thedifference between the meanswas
Transepithelialpotentials highlysignificant(P<0.001, two-tailedttest). Thesodium
concentration ofthe medium was 51 mA/and that ofthe
The values oftheTEP's in the filamentswere indepen- hemolymph was 107.5 15.0 mA/ (mean S.D.,
=
dent ofthe region (proximal, medial, or distal) ofthe fil- n 11).
GILL FILAMENTS IN P. CLARKII 163
Thestructure ofthethin squamousepithelia in P. clar-
kiirespiratory filamentswassimilartothe respiratory ep-
itheliadescribed in othercrayfish (Dunel-Erbetai, 1982)
in crabs (Copeland and Fitzjarrell, 1968; Taylor and
Greenaway. 1979), and in shrimp (Foster and Howse,
/^.rnol P 1978). The paucity oforganelles observed in the epithe-
mg protein*hr lium and pillar structures indicates that little cellular ac-
tivity occurs in these cells other than maintenance ofcell
components. The thin epithelial layer mayserveasaper-
meability barrier to the diffusive loss ofions and blood
proteins, while the thin cytoplasm would allow efficient
gas exchange to occur.
Large bundles of microtubules were observed in the
Total Na.K pillarstructuresofthe respiratory filaments,whereassingle
Figure20. Mean( standarddeviation)ATPaseactivitycomparing microtubules were the rule in the transporting epithelia
pooled transporting (open bars) and respiratory (solid bars) filaments. ofthis crayfish and in most gill epithelia ofother crusta-
** = P<0.001 ceans (Copeland and Fitzjarrell, 1968; Bielawski, 1971;
Foster and Howse, 1978; Taylor and Greenaway, 1979;
Finol and Croghan, 1983; Compere et ai, 1989). These
Discussion microtubules appear to anchor the epithelium to the cu-
ticle. Finol and Croghan (1983) have proposed that the
The results ofsilver staining, ultrastructural analyses, microtubules function to stabilize the gill epithelium
ATPase measurements, and measurements ofTEPall in- against the hydrostatic pressure ofthe blood. Becausethe
dicate that the gills ofP. clarkii are not homogeneous in respiratory epithelium contains very little cytoplasm, ad-
structure, but contain filamentsdedicated to ion transport ditional microtubules, in the form of bundles near the
and filaments dedicated to respiration. No filaments had periphery ofthe pillarstructures, may give theadditional
both characteristics; rather individual filaments were support needed to withstand the shearforcesofthe blood
uniquely respiratoryortransporting. Respiratory filaments flow in these filaments.
occurred on the lateral rows of most gills, whereas the The loose connective tissue in the gills ofP. clarkii is
transporting filaments predominated in the central bed similar to that observed in Callinectes sapidus (Johnson,
ofthe gills. This distribution ofthe two filament types in 1980). The cells within the loose connective tissue, with
discrete regionsofthegill placed the respiratory filaments their abundant glycogen rosettes and granules, may reg-
in the path ofthe most rapid water flow (Burggren el ai, ulate the blood glucose levels as suggested by Finol and
1974). Croghan 1983) for Uca more/a.
(
The respiratory filaments had a very thin, squamous
epithelium, whereas the ion transporting filaments pos-
sessed a markedly thicker epithelium. The ultrastructure
ofthe ion transportingepithelia observed in P. clarkiigill
filaments was similar to ion transporting epithelia de-
scribed in other crayfish (Morse el a/., 1970; Bielawski.
1971; Fisher, 1972), in isopods (Bubel and Jones, 1974),
in crabs (Copeland and Fitzjarrell, 1968; Taylor and
Greenaway, 1979; Finol and Croghan, 1983), in shrimp
(Foster and Howse, 1978), and in Dapluua (Kikuchi, Frequency
1983). Apical microvillar processes were present to in-
crease the surface area over which ions could be trans-
portedaswellasnumerousbasal infoldingswithassociated
mitochondria. Such basal infoldings are reported to be
the major site ofNa, K-ATPase (Diamond and Bossert,
1968; Ernst, 1972; Ernst et ai. 1981; Towle, 1985;Towle
and Kays, 1986). Thiswould be consistent with the higher -16 -24 -32 -40
Na, K-ATPase activity measured in transporting fila- T.E.P (mV)
ments, becausefewbasal infoldingsormitochondriawere Figure21. Transepithelialpotentials(TEP)forrespiratory(solidbars,
observed in the epithelia ofthe respiratory filaments. n = 19)and transporting(open bars, n = 21) filaments.