somero 2002_zonacao temperatura

780

INTEG. AND COMP. BIOL., 42:780–789 (2002)

Thermal Physiology and Vertical Zonation of Intertidal Animals:Optima, Limits, and Costs of Living1

GEORGE N. SOMERO2

Hopkins Marine Station, Stanford University, Pacific Grove, California 93950-3094

SYNOPSIS. Temperature’s pervasive effects on physiological systems are reflected in the suite of tempera-ture-adaptive differences observed among species from different thermal niches, such as species with dif-ferent vertical distributions (zonations) along the subtidal to intertidal gradient. Among the physiologicaltraits that exhibit adaptive variation related to vertical zonation are whole organism thermal tolerance,heart function, mitochondrial respiration, membrane static order (fluidity), action potential generation, pro-tein synthesis, heat-shock protein expression, and protein thermal stability. For some, but not all, of thesethermally sensitive traits acclimatization leads to adaptive shifts in thermal optima and limits. The costsassociated with repairing thermal damage and adapting systems through acclimatization may contributeimportantly to energy budgets. These costs arise from such sources as: (i) activation and operation of theheat-shock response, (ii) replacement of denatured proteins that have been removed through proteolysis,(iii) restructuring of cellular membranes (‘‘homeoviscous’’ adaptation), and (iv) pervasive shifts in geneexpression (as gauged by using DNA microarray techniques). The vertical zonation observed in rocky in-tertidal habitats thus may reflect two distinct yet closely related aspects of thermal physiology: (i) intrinsicinterspecific differences in temperature sensitivities of physiological systems, which establish thermal optimaand tolerance limits for species; and (ii) ‘cost of living’ considerations arising from sub-lethal perturbationof these physiological systems, which may establish an energetics-based limitation to the maximal height atwhich a species can occur. Quantifying the energetic costs arising from heat stress represents an importantchallenge for future investigations.

INTRODUCTION

The latitudinal and vertical patterning of species’distributions commonly reflects gradients or disconti-nuities in environmental temperature. For example,along the subtidal to intertidal gradient, species re-placements are common, and there are many examplesof one species in a genus replacing a congener overrelatively small vertical ranges where thermal gradi-ents exist (Gowanlock and Hayes, 1926; Broekhuysen,1940; Evans, 1948; Wolcott, 1973; Newell, 1979;Garrity, 1984; Tomanek and Somero, 1999; Stillmanand Somero, 2000). Because of the pervasive effectsof temperature on physiological systems, the tight link-age observed between environmental temperature andspecies’ distribution patterns is to be expected, espe-cially for ectothermic species whose body tempera-tures may vary widely on seasonal and tidal cycle timescales (Helmuth, 1998, 1999). Virtually all biologicalstructures and processes are affected by temperature,and, for this reason, temperature-adaptive variationthat establishes different thermal optimal and thermallimits among species is widespread (Hochachka andSomero, 2002). Much of our initial knowledge aboutthis adaptive variation came from studies of speciesthat are widely divergent phylogenetically. Althoughstudies of widely diverged species have revealed majoradaptive patterns, they generally have not provided ev-idence concerning the role of adaptation to tempera-

1 From the Symposium Physiological Ecology of Rocky IntertidalOrganisms: From Molecules to Ecosystems presented at the AnnualMeeting of the Society for Comparative and Integrative Biology, 2–7 January 2002, at Anaheim, California.

2 E-mail: [email protected]

ture in establishing fine-scale patterns of zonation,such as that seen along the subtidal-intertidal gradient.To address the latter question, it has proven useful tocompare closely related species, especially congenersthat exhibit temperature-related patterns of zonation.Studies of congeners allow adaptive variation to bedelineated clearly, independent of effects of phylogeny(see Stillman and Somero, 2000). The chief objectiveof this review is to illustrate how the study of conge-neric species found at different tidal heights along thesubtidal-to-intertidal axis has provided especially clearinsights into the roles of temperature adaptation infine-scale spatial patterning, including the establish-ment of the upper limits of distribution in rocky inter-tidal habitats.

In this review of the thermal physiology of rockyintertidal organisms, two distinct, yet closely related,issues related to the role of temperature in establishingvertical zonation will be addressed. First, I examinetemperature-adaptive physiological differences amongspecies, which are instrumental in establishing theirthermal optima and thermal tolerance limits and, there-by, their different vertical zonations. A variety of phys-iological systems are examined, including systems thatare genetically fixed and not subject to acclimatization(seasonal or diurnal) and others that are phenotypicallyplastic and can be adaptively modified during the in-dividual’s lifetime, in some cases during the course ofa single tidal cycle (Williams and Somero, 1996). Sec-ond, I consider how thermal damage to temperature-sensitive structures and processes, followed by its re-pair, and the acclimatization of phenotypically plasticsystems affect the ‘cost of living’ faced by organisms

781TEMPERATURE AND VERTICAL ZONATION

FIG. 1. Differences in whole organism heat tolerance between con-generic species found at different heights along the subtidal to in-tertidal gradient. Upper panel: temperatures at which 50 percent le-thality occurred (LT50) for 19 species of porcelain crabs (genus Pe-trolisthes) (Stillman and Somero, 2000). The dashed line labeledLT50 5 MHT is a line for which lethal temperature equals the max-imal habitat temperature (MHT). Middle panel: effects of heating ata rate of 18C per 12 min on survival of congeners of Tegula (To-manek and Somero, 1999). Lower panel: effects of thermal exposure(3 hr of exposure under immersion in seawater, followed by 45 minof recovery) on survival of congeners of Littorina (M. Phillips andG. N. Somero, unpublished data).

with different vertical zonations. I show that, whereaswe have made considerable progress in delineating theadaptive modifications of physiological, biochemical,and molecular traits that may be responsible for estab-lishing thermal optima and thermal limits, a high levelof uncertainty remains about the role that thermal ef-fects on bioenergetics plays in vertical patterning. Al-though it is not possible at present to quantify the bio-energetic costs of thermal stress, there is increasingevidence that these costs play important roles in estab-lishing patterns of zonation. Thus, the long-standinghypothesis that predicts a pivotal role for abiotic fac-tors like temperature in restricting upper distributionslimits (see Connell, 1961) will be seen to be supportedby many of the recent findings discussed in this re-view.

In my analysis, I will take a reductionist approachto illustrate how differences observed at the levels ofthe whole organism, organ function, cellular activity,organelle function, biochemical pathways, and molec-ular-level phenomena all reflect temperature-adaptivedifferences among congeneric species. This review ofadaptive variation is designed to at once show thescope of temperature-adaptive change and pinpointmany of the sites of thermal perturbation that are likelyto incur substantial costs in energy for repair or adap-tive modification. As indicated above, I will focus oncongeneric species that differ in their vertical distri-butions and allow an especially clear analysis of adap-tive variation related to temperature. Importantly forthe goal of developing general principles about ther-mal effects and distribution patterns, the many casesin which replacement of one congener by another isfound across the subtidal to intertidal gradient allowone to conduct a taxonomically broad analysis. Thisreview will examine congeneric groups of snails, mus-sels, crabs, abalone, and fish, in an effort to reach gen-eral conclusions that may be applicable across the an-imal kingdom and, for many ubiquitous physiologicaltraits, for other types of organisms as well.

WHOLE ORGANISM THERMAL TOLERANCE CORRELATES

WITH VERTICAL POSITION ALONG THE SUBTIDAL TO

INTERTIDAL GRADIENT

Measuring heat death of animals is technically quitesimple. However, because the intensity of heat stressis a function of several variables, including the pastthermal history of the specimens, the magnitude of theincrease in temperature, the rate at which heating oc-curs, and the duration of exposure to heat, valid com-parisons among studies often are difficult if not im-possible to make. However, when consistent experi-mental treatments are employed, striking patterns re-lating adaptation (or acclimation) temperature and heattolerance are observed, as has been emphasized byother contributors to this symposium (Stillman, 2002;Tomanek, 2002). Figure 1 presents data on threegroups of marine invertebrates for which congener re-placement patterns along the subtidal to intertidal gra-dient are found. In all cases, there is a strong corre-

782 GEORGE N. SOMERO

FIG. 2. Thermal limits of heart rate in Petrolisthes cinctipes and P.eriomerus. Temperatures of immobilized crabs were increased at arate of 18C per 15 min, a heating rate that mimics heating underextreme field conditions. Responses of single individuals of eachspecies are shown. Arrhenius break temperature (ABT) values areaverages of 6 individuals for each species. (Figure modified afterStillman and Somero, 1996).

lation between increasing tidal height and resistance tohigh temperature. For porcelain crabs (genus Petrol-isthes), 19 species from different latitudes and verticalpositions show a highly significant correlation betweenupper lethal temperature (LT50, the temperature atwhich 50% mortality occurs) and maximal habitattemperature (MHT), a trend that is independent ofphylogenetic influences (Stillman and Somero, 2000)(Fig. 1A). As shown by the line of equality betweenMHT and LT50, those species of porcelain crabs thatcurrently encounter the highest habitat temperaturesgenerally are living nearer to their LT50 than congenersfound in cooler habitats. Even in the case of a northtemperate mid-intertidal species, Petrolisthes cinctipes,MHT approaches LT50 (see below). Species for whichLT50 is close to MHT may be especially vulnerable toeffects of global warming. The congruence betweenMHT and LT50 for several species of porcelain crabscontradicts earlier conjectures that intertidal animals,although differing in thermal tolerance in relation totheir vertical zonation, do not encounter lethal thermalconditions under natural conditions (see Wolcott,1973). In the context of physiological determinants ofvertical zonation, note that many of the more cold-adapted porcelain crabs could not withstand the habitattemperatures experienced by more warm-adapted con-geners. For instance, P. eriomerus, a low-intertidal tosubtidal species whose latitudinal distribution overlapsthat of P. cinctipes, dies at temperatures several de-grees below the MHT of the latter, higher-occurringspecies.

For turban snails (genus Tegula) differences are ob-served between two subtidal to low-intertidal species(T. montereyi and T. brunnea) and a mid-intertidal spe-cies, T. funebralis, which have overlapping latitudinal

distributions (Tomanek and Somero, 1999) (Fig. 1B).As in the case of the two species of porcelain crabsjust discussed, the two subtidal to low-intertidal spe-cies of Tegula are unable to withstand the habitat tem-peratures that may be experienced by their mid-inter-tidal relative during extremes of heat stress. For litto-rinid snails, which commonly are one of the highest-occurring intertidal invertebrates, thermal toleranceagain is correlated with vertical position (Fig. 1C). Thedifferences in thermal tolerance seen among thesethree sets of congeners, in conjunction with earlierstudies that have shown correlations between verticalposition and heat tolerance (Gowanlock and Hayes,1926; Evans, 1948; Wolcott, 1973; Garrity, 1984), pro-vide a strong basis for assigning an important role totemperature-adaptive physiological variation in estab-lishing patterns of vertical zonation. Can we identifythe temperature-sensitive physiological systems thatunderlie the whole organism effects just discussed andthereby explain mechanistically the strong correlationbetween thermal tolerance and zonation and the basisof heat death, especially in cases where LT50 and MHTclosely coincide?

THERMAL LIMITS OF PHYSIOLOGICAL SYSTEMS: AMULTI-LEVEL SEARCH FOR ‘‘WEAK LINKS’’

In view of the pervasive nature of thermal effectson biological structures and processes, there is a stronga priori basis for arguing that all links in the physio-logical chain are apt to be ‘‘weak,’’ although somesystems may prove more vulnerable to thermal stressthan others (Hochachka and Somero, 2002). Surveysof diverse physiological systems, in a taxonomicallyvaried range of species, have shown that this predic-tion is correct. Below, I present selected examples ofthermally sensitive physiological systems, which offerespecially clear cases of how thermal physiology con-tributes to patterns of zonation in rocky intertidal hab-itats. In some cases, the data help to explain mecha-nisms of acute heat death; in other cases, insights arepresented concerning how sub-lethal effects of hightemperature may lead to substantial increases in theorganism’s ‘‘cost of living.’’

Organ-level effects: heart function in porcelain crabs

The first example of temperature-adaptive variationamong homologous physiological systems of conge-ners with different vertical positions involves the effectof acute temperature change on heart rate in two con-geners of Petrolisthes. As shown in Figure 1A, P.cinctipes and P. eriomerus differ in LT50 by 58C(32.38C and 27.58C, respectively), in accord with theirdifferent MHTs (approximately 338C and 168C, re-spectively) (Stillman and Somero, 1996, 2000). Theresponses of the hearts of these two congeners to risingexperimental temperature differ in manners that sug-gest that heart function plays an important role in set-ting zonation patterns. For the less thermally tolerantspecies, P. eriomerus, heart rate falls dramatically at26.68C (Fig. 2). This is the Arrhenius ‘‘break’’ tem-


FIG. 3. Effects of incubation at 308C on action potential generationin an abdominal mechanosensory nerve bundle for four congenersof Petrolisthes. P. armatus is an intertidal species found through thetropical Pacific; P. cinctipes is a cool temperate low- to mid-inter-tidal species; and P. eriomerus and P. manimaculis are cool tem-perate subtidal species. The number of action potentials per min arestandardized to the number observed at the initiation of the 308Cexposure (Jessica Knape and GNS, unpublished data).

perature (ABT), the temperature at which a sharp dis-continuity in slope occurs in an Arrhenius plot (lograte vs reciprocal of absolute temperature (K)). For themore heat-tolerant P. cinctipes, the ABT is several de-grees higher (31.58C). These data support three im-portant conclusions about the role of physiologicalsystems in vertical zonation. First, heart function ofthe lower-occurring P. eriomerus could not be sus-tained at the habitat temperatures encountered duringhot periods by P. cinctipes. Thus, P. eriomerus couldnot invade the habitat of its higher-occurring congener.Second, for P. cinctipes, the ABT of heart function isessentially identical with the LT50. Thus, heart failuremay be one of the proximate causes of heat death inthis species. Third, the close proximity of upper habitattemperature to the upper thermal limit of heart func-tion suggests that this species may be living on theedge of ‘‘heart failure’’ in its native habitat. It remainsto be established how plastic the ABT of heart func-tion is during acclimatization. Whole organism ther-mal tolerance studies of crabs acclimated to 88C and188C revealed that the LT50 increased by 48C for P.eriomerus and 28C for P. cinctipes (Stillman and So-mero, 2000). Thus, crab hearts may be able to adjusttheir thermal tolerance (ABT) to at least some degree.Note that the more limited ability to warm acclimateobserved for P. cinctipes is further evidence that thisspecies lives near the limits of its thermal tolerance insitu. It is noteworthy that, even though P. cinctipes ismore eurythermal than P. eriomerus, as shown by itsgreater ability to survive at low as well as at hightemperatures (Stillman and Somero, 1996), it has alower capacity for thermal acclimation to high tem-peratures than its more stenothermal congener.

Differences between these two congeners of Petrol-isthes also were observed in cold tolerance of heartfunction: heart function of P. cinctipes is both moreheat tolerant and more cold tolerant than that of P.eriomerus (Stillman and Somero, 1996). Thus, eventhough this review focuses primarily on heat stress, itis important to emphasize that tolerance of low tem-perature may also be important in establishing zona-tion along the subtidal-intertidal gradient. Species withthe highest vertical distributions are apt to be moreeurythermal than low-intertidal and subtidal species.As will be shown later in the contexts of protein ther-mal stability and membrane structure, the need for amid- to high-intertidal species to function over a rel-atively wide thermal range may incur considerablephysiological costs, which may serve to establish theupper limits of these species’ distributions.

Cellular level effects: action potential generation

The nervous systems of animals are known to beespecially sensitive to temperature change and to serveas critical sites in establishing upper and lower thermallimits (Cossins et al., 1977; Cossins and Bowler, 1987;Hochachka and Somero, 2002). Although few data ex-ist on the thermal sensitivities of neural function indifferently adapted congeneric species found across

the subtidal-intertidal gradient, what information isavailable suggests the presence of significant adaptivevariation related to habitat temperature. Recent studiesof porcelain crabs have revealed interspecific variationthat closely mirrors the differences seen for whole an-imal heat tolerance (Fig. 1A) and heart function (Fig.2). One illustration of these effects is the influence ofincubation temperature on rates of action potentialgeneration in a mechanosensory nerve bundle (Fig. 3;Knape and Somero, unpublished observations; also seeStillman, 2002). At an experimental temperature of308C, the mechanosensory bundle of the subtropicalspecies P. armatus shows stable firing over a longerperiod than was observed for the temperate mid-inter-tidal species P. cinctipes. In turn, P. cinctipes is lessheat-sensitive than two subtidal to low-intertidal con-geners, P. eriomerus and P. manimaculis, whose lati-tudinal distributions overlap with P. cinctipes. Al-though these initial studies of nerve function in dif-ferently thermally adapted congeners do not allow de-finitive conclusions to be drawn about the thermallimits of neural function, the trends shown in Figure3 do provide a basis for concluding that temperature-adaptive variation in thermal sensitivities of nervoussystems may contribute importantly to vertical zona-tion.

Organelle-level effects: mitochondrial respiration

Like the rate of heart beat, the rate of oxygen con-sumption by isolated mitochondria typically exhibits a‘‘break’’ at some elevated temperature, the ABT, whenvalues are displayed on Arrhenius plots. Figure 4 il-lustrates how the ABT of mitochondrial oxygen con-sumption varies with adaptation and acclimation tem-perature among a wide range of marine fishes and in-vertebrates (mollusks, annelids, and arthropods) (Dahl-


FIG. 4. Arrhenius break temperatures of respiration by mitochon-dria of diverse marine invertebrates and fishes. The data for theAntarctic fish Trematomus bernacchii are not included in the re-gression analysis. The dotted line is a line of equality between ABTand maximal habitat temperature. (Figure modified after Weinsteinand Somero, 1998)

FIG. 5. (A) The effects of acclimation temperature on the Arrheniusbreak temperature of mitochondrial respiration for four species ofabalone (genus Haliotis). Values are means 6 1 S.E.M. N 5 4 forall species and temperatures except for pink abalone at 238C andpinto abalone at 5C, where N 5 3. Overlapping data points (red andpink @ 208C and pinto and red @ 58C) have been offset by 18C forclarity. (B) The effects of acclimation temperature on mitochondrialmembrane fluidity, as measured by the fluorescence polarization ofthe probe 1,6-diphenyl-1,3,5-hexatriene (DPH) for congeners of ab-alone. N 5 3 for all species. The eurythermal green abalone (H.fulgens) occurs intertidally and encounters temperatures of 14–278Cover its biogeographic range. The pink abalone (H. corregata) oc-curs subtidally and encounters temperatures of 12–238C over its bio-geographic range. The red abalone (H. rufesens) is a cool temperatespecies with a habitat temperature range of 8–188C. The pinto aba-lone (H. kamtschatkana) is the most northern-occurring species andencounters habitat temperatures between 4 and 148C. See Dahlhoffand Somero (1993) for additional data on the distributions and ther-mal regimes of these species. (Figure modified after Dahlhoff andSomero, 1993)

hoff and Somero, 1993; Weinstein and Somero, 1998).ABT increases with adaptation temperature amongspecies and with acclimation temperature in single spe-cies (see below). However, the rise in ABT with in-creasing adaptation or acclimation temperature has aslope of less than unity, as shown by comparison ofthe regression line fit to available ABT data (solid line;the value for the Antarctic fish is not included in theregression analysis) and the line for which ABT 5adaptation temperature (dashed line). The higher thetemperature of adaptation, the nearer is the ABT ofmitochondrial respiration to the species’ adaptationtemperature. The two lines intersect at a temperaturebetween 55 and 608C. This intersection suggests that,regardless of adaptation temperature, mitochondrialfunction cannot increase its thermal stability, as in-dexed by the ABT of respiration, above 55–608C.These temperatures, then, may be the highest temper-atures to which animals can adapt (Hochachka and So-mero, 2002).

Within a species’ normal (5physiological) range ofbody temperatures, the ABT of mitochondrial respi-ration may be subject to adaptive modification throughacclimatization or acclimation. Phenotypic plasticity inthe ABT of mitochondrial respiration has been ob-served in congeners of abalone (genus Haliotis) fromdifferent thermal habitats (Dahlhoff and Somero,1993). Figure 5A illustrates variation in ABT amongdifferently thermally acclimated individuals of fourcongeners that occur at different latitudes and at dif-ferent vertical positions along the subtidal-to-intertidalgradient (see legend of Fig. 5 for distributions). Whencomparisons are made between field-acclimatizedspecimens (data not shown; see Dahlhoff and Somero,1993), relatively warm-adapted species that occur intothe intertidal zone, e.g., green (H. fulgens) and black(H. cracherodii) abalone, have higher ABTs than more

cold-adapted species, e.g., the pinto abalone (H. kam-tschatkana), which is the northernmost species studiedand which is found intertidally only at high latitudes(Gulf of Alaska and northern British Columbia). How-ever, these interspecific differences are not fixed; allspecies examined (inadequate availability of black ab-alone precluded use of this species in acclimation stud-ies) showed acclimatory plasticity in ABT. Following


FIG. 6. Induction of heat-shock protein 70 (hsp70) synthesis in fourcongeners of Tegula acclimated to 238C. Gill tissue was incubatedat a series of temperatures in seawater containing 35S-labeled me-thionine and cysteine. Proteins were resolved by SDS-PAGE andradioactivity was quantified by autoradiographic and densitometricanalysis (see Tomanek and Somero (1999) for protocols). Theamounts of hsp70 synthesized during the incubation and recoveryperiod are normalized to the levels of hsp70 synthesis observed intissue exposed to 138C. Values are means 6 1. S.E.M. N 5 5 forall data points except for T. funebralis @ 338C (N 5 4) and T.brunnea @ 138C. (Figure modified after Tomanek and Somero,1999).

acclimation to a common temperature, ABT valueswere similar among species. However, the thermalrange over which ABT was modified was typicallywider for more eurythermal species than for steno-thermal species (compare green and pinto abalones).

Membrane-level effects: Alterations in bilayer staticorder

To determine the mechanistic basis of the observeddifferences in ABT among the differently acclimatedabalone, the static order of the mitochondrial mem-branes was measured, using the technique of fluores-cence polarization (Dahlhoff and Somero, 1993). Thistechnique involves intercalation into the membrane bi-layer of a fluorescent probe molecule such as 1,6-di-phenyl 1,3,5-hexatriene (DPH). The static order of thebilayer determines the mobility of the probe, such thatthe strength of the polarization signal is directly relatedto the rigidity of probe’s microenvironment. A morerigid (less fluid) microenvironment prevents the probefrom moving freely, thereby enhancing the probe’sability to emit polarized light in an orientation that canbe detected by the fluorescence spectrophotometer.

As shown in Figure 5B, there was a generally sim-ilar pattern in the responses of ABT and DPH polari-zation intensity to acclimation in the four species ofabalone. Rising acclimation temperature led to increas-es in ABT and generally parallel increases in the ri-gidity of the membrane lipid bilayer (as shown bystronger DPH polarization signals). The term ‘‘hom-eoviscous adaptation’’ is commonly used to refer tothe stability in membrane static order that is acquiredduring adaptation or acclimation to temperature (seeHazel, 1995). Broad interspecific comparisons, as wellas studies of differently acclimated specimens, haveshown that membrane static order is strongly con-served at normal body temperatures. There appears tobe an optimal range of membrane fluidity and phasethat allows such critical membrane functions as activetransport, enzymatic activity, and exocytosis to be con-served at physiological temperatures (Hazel, 1995).

The changes in membrane static order, the homeov-iscous adaptations, observed in mitochondria from dif-ferently acclimated abalone were likely the conse-quence of several types of alterations in membrane lip-id composition. Comparisons of membrane lipids fromanimals adapted and acclimated to different tempera-tures have revealed pervasive changes in head group,acyl chain saturation, and chain length (see Hazel,1995; Logue et al., 2000). Alterations in membranelipid composition may be the most common type ofadaptation to temperature (Hazel, 1995).

In the case of rocky intertidal invertebrates, whichmay encounter changes in body temperature in excessof 258C during a single tidal cycle (see Helmuth, 1998,1999; Tomanek and Somero, 1999), rapid adjustmentsin membrane lipid composition may be needed tomaintain the appropriate membrane static order. In-deed, studies of Mytilus californianus showed that, inaddition to pronounced seasonal time-scale homeovis-

cous adaptation, mussels were able to alter their mem-brane lipid composition in response to temperaturechanges associated with a tidal cycle (Williams andSomero, 1996). The ability to alter membrane staticorder on time scales of hours was found in musselsfrom the high intertidal zone but not in conspecificsthat occurred at low intertidal sites. These differencesamong conspecifics suggest that the exposure historyof an organism may affect its ability to respond toenvironmental change, an issue that is addressed in thenext section as well as in the concluding discussion ofthis review.

Biochemical systems: Protein synthesis and the heat-shock response

Variation among congeners in thermal optima andtolerance limits also has been observed in protein bio-synthesis. For example, congeners of Tegula differ inheat tolerance of protein synthesis and in the pattern-ing of heat-shock protein expression in response toheat stress (Tomanek, 2002; Tomanek and Somero,1999, 2000, 2002). Congeners of the mussel genusMytilus likewise have shown differences of these types(Hofmann and Somero, 1996a). As shown in Figure6, 238C-acclimated congeners of Tegula induce ex-pression of heat-shock protein-70 (hsp70) at tempera-tures that reflect the species’ widely different adapta-tion temperatures. The most warm-adapted speciesstudied, T. rugosa, which occurs intertidally in theGulf of California, did not induce hsp70 synthesis until308C and exhibited maximal induction at 368C. Syn-thesis of hsp70 ceased by 428C. At the other extreme,two temperate zone subtidal to low intertidalspecies,T. montereyi and T. brunnea, induced hsp70 at 278C


FIG. 7. Thermal stabilities of malate dehydrogenases (MDH) ofcongeners of Littorina and Tegula. Homogenates of foot tissue wereprepared in 20 mM imidazole-Cl buffer (pH 7.0 @ 58C) containing1 mM EDTA and 5 mM DTT. Homogenates were centrifuged at15,800 g for 30 min and supernatants were used for activity assays.Residual activity at various time points during incubation at 408Cwas measured in a reaction mixture containing 80 mM imidazole-Cl buffer (pH 7.0 @ 208C), 100 mM KCl, 150 mM NADH and 200mM oxaloacetic acid. Loss of activity during incubation at 408C wascorrected for any loss of activity in a control sample kept on ice.(Data of Dr. Lars Tomanek (Tegula) and Michelle Phillips and theauthor (Littorina))

and showed maximal synthesis at 308C. Synthesis ofhsp70 ceased by 338C. Although not shown in Figure6, the maximal temperature of synthesis of hsp70 co-incides with the upper temperature at which proteinsynthesis per se is possible in the four congeners (To-manek and Somero, 1999). Temperate zone subtidal tolow intertidal species of Tegula (T. montereyi and T.brunnea) are unable to synthesize proteins at temper-atures that the mid-intertidal T. funebralis may en-counter during emersion on hot, clear days. In turn, T.funebralis is unable to synthesize proteins at temper-atures encountered by the subtropical species, T. ru-gosa.

One important difference between the thermal ef-fects encountered by subtidal and intertidal species liesin the frequency with which the heat-shock responseis likely to be induced. For T. brunnea and T. mon-tereyi, hsp70 induction temperatures are several de-grees higher than the maximal habitat temperaturesthese two subtidal species are likely to experience (To-manek and Somero, 1999). In contrast, both mid-in-tertidal species studied, and T. rugosa in particular, arelikely to experience regularly thermal conditions thatactivate hsp70 synthesis. These differences in frequen-cy and intensity of synthesis of heat-shock proteinscould result in significant differences in ‘cost of living’between subtidal and intertidal animals, as discussedbelow.

Protein-level effects: Enzyme thermal stability

The activation of the heat-shock response is indic-ative of thermal denaturation of proteins during peri-ods of heat stress (Feder and Hofmann, 2000). Theobservation that the temperatures at which the heat-shock response is activated correlate positively withadaptation temperature, as shown in Figure 6 for con-geners of Tegula, suggests that species adapted to dif-ferent temperatures have proteins with different intrin-sic thermal stabilities. Such a trend between evolu-tionary adaptation temperature and intrinsic stability oforthologous proteins is common (see Hochachka andSomero, 2002, for review) and is illustrated by malatedehydrogenases (MDHs) from five species of Tegulaand Littorina (Fig. 7). Thermal stabilities of malatedehydrogenase (MDH) increase regularly with increas-ing vertical position. Differences in protein thermalstability of the type observed for the MDHs of thesemolluscs thus may account in part for different onsettemperatures of heat-shock protein expression.

The observation that proteins may denature at tem-peratures lying within the upper reaches of the phys-iological temperature range raises an interesting ques-tion about what might be regarded as an apparent‘‘shortfall’’ of evolution. Proteins can be modified dur-ing evolution and, now, in the molecular biologist’slaboratory to attain extremely high levels of intrinsicstability. Most remarkable are proteins from thermo-philic members of the Domain Archaea, which are sta-ble at temperatures in excess of 1008C (reviewed inHochachka and Somero, 2002). Would it not be ben-

eficial to an intertidal invertebrate to possess proteinsthat are sufficiently rigid to avoid denaturation even atthe highest body temperatures that are experiencedduring emersion on hot, sunny days?

The answer to this question involves an integrativeperspective on protein structure and function, one inwhich the evolution of a single characteristic such asthermal stability is weighed against possible negativeeffects on other characteristics of the protein. In thecase of enzymatic proteins, which generally alter theirconformations during function, it is necessary to main-tain an optimal balance between rigidity of structure,which allows persistence of the native folded state, andflexibility, which ensures that the conformationalchanges required for function can occur at appropriaterates at normal cell temperatures. In essence, this struc-ture-function relationship dictates that the ‘‘movingparts’’ of proteins must reach a balance between sta-bility and flexibility that is appropriate for the normaltemperatures at which the proteins work (Fields andSomero, 1998). Proteins of cold-adapted species thushave a higher degree of conformational flexibility thanproteins of warm-adapted species. In the case of eu-rythermal intertidal species, which face wide variationin temperature during alternate periods of immersionand emersion, enzymatic function must be retainedover a wide range of cell temperatures. In fact, as dis-cussed later, metabolic processes may be especially ac-tive during periods of immersion, when access to ox-ygen and food is greater than under conditions ofemersion. Therefore, if a mid-intertidal species like T.funebralis were to evolve highly rigid enzymes to


avoid thermal unfolding at the highest temperatures itexperiences during emersion, metabolic activity at thelow temperatures characteristic of immersion might becompromised because these relatively rigid proteinswould be incapable of undergoing rapid reversiblechanges in conformation in the cold. The need to haveenzymes and other proteins that can function well atlow temperatures during immersion requires that heat-tolerant, eurythermal species like T. funebralis face theconsequences of thermal denaturation in situ. The re-pair and replacement of heat-damaged proteins thusmay represent a significant and unavoidable energeticcost for intertidal species.

‘COST OF LIVING’ ISSUES: DO COSTS OF REPAIR AND

REBUILDING LIMIT VERTICAL POSITION?

Marine ecologists have long regarded the stressesimposed by abiotic factors like temperature to playpivotal roles in establishing the upper limits to zona-tion in the intertidal zone (Connell, 1961; Newell,1979; Newell and Branch, 1980). The foregoing sec-tions that described the effects of temperature on adiverse suite of physiological traits provide strong sup-port for this hypothesis. Although it is impossible atpresent to affix a firm number, e.g., a percentage oftotal metabolic turnover, to the costs of coping withthermal stress, the data presented above do imply thatan intertidal species that experiences extensive periodsof emersion is apt to spend a considerable amount ofmetabolic effort to repair, replace, and restructure ther-mally sensitive biochemical components of the cell.

Protein replacement and repair

One of the best-understood aspects of thermal per-turbation involves the damage done to proteins and thesubsequent restoration of protein homeostasis. Theheat-shock response may be strongly activated at phys-iological temperatures, at least in the case of intertidalspecies like T. funebralis and T. rugosa (Fig. 6) andmussels of the genus Mytilus (Hofmann and Somero,1995, 1996a). Energy is required at several events inthe heat-shock response, including the activation oftranscription of heat-shock genes, the synthesis ofhsps, and the ATP-requiring chaperoning by hsps. Fur-ther energy is required if proteins are irreversibly de-natured and cannot be rescued by heat-shock proteins.Many denatured proteins undergo a process known asubiquitination in which a small protein, ubiquitin, isbound covalently to the denatured target protein. Ubi-quitination requires ATP, as does the subsequent pro-teolysis of the ubiquitinated protein through non-ly-sosomal proteolytic pathways. Heat-stressed mussels(M. trossulus) found in high-intertidal habitats con-tained significantly higher levels of ubiquitinated pro-teins than subtidal conspecifics (Hofmann and Somero,1995), another indication that vertical position affectsthe energetic costs of life. Proteins that are degradedmust, of course, be replaced, if metabolic activities areto be sustained. The energy-demanding process of pro-tein biosynthesis, which may require synthesis of the

machinery for biosynthesis, e.g., ribosomal proteins,as well as synthesis of replacement proteins, thus mayfigure importantly in calculations of energy costs ofheat stress.

Membrane-level effects: Temperature-dependentpermeabilities and lipid restructuring to maintainstatic order

Two types of energetic costs can be envisioned interms of the thermal effects on membrane function andstructure that an intertidal species would face duringalternate periods of emersion and immersion. Increasesin temperature during emersion will tend to increasethe passive permeability of membranes, thereby lead-ing to a rise in uncontrolled flux of small ions andmolecules (including water) through the membrane.Restoration of trans-membrane gradients will requireenergy, e.g., more activity of the Na1-K1-ATPase (so-dium pump). The restructuring of the membrane bi-layer, which may occur during the tidal cycle, if bodytemperature fluctuates widely during emersion and im-mersion, represents a second energetic cost to intertid-al species. Estimating the costs entailed in maintainingan appropriate static order is not currently possible andremains a challenge for future investigation.

Anaerobic metabolism: Another cost to intertidalspecies

Membrane-localized effects linked to emersion maylead to another important impact on energy budgets.Loss of cellular water across membranes may accom-pany emersion, especially if temperatures and wind ve-locities are high, adding a threat of desiccation to thecell. This threat may be met by a reduction in gasexchange across respiratory surfaces, as animals se-quester these surfaces from the air during emersion.This response is common, as seen in the closures ofmussel valves and opercular openings in snails duringemersion. A consequence of this reduction in the threatof desiccation may be a shift away from efficient ox-ygen-based pathways of ATP generation to less effi-cient anaerobic ATP generating reactions, which arehighly developed in many intertidal invertebrates(DeZwann and Mathieu, 1992).

If an invertebrate does undergo a shift towards an-aerobic ATP generating pathways during emersion, itmay only be during the subsequent period of re-im-mersion in oxygen-rich water that it will be able togenerate sufficient ATP to support the costs of proteinrepair and replacement. In studies of the mussel My-tilus californianus, it was found that activation of theheat-shock response and ubiquitination of heat-dam-aged proteins occurred primarily after the animalswere re-immersed following heat stress (Hofmann andSomero, 1996b). Thus, under conditions of heat stress,the longer the period of emersion relative to the timeof immersion, the greater will be the amount of heatdamage and the more limited will be the time for re-dressing this injury. Furthermore, if feeding is reducedduring emersion, the energetic costs linked to repair of


FIG. 8. Growth rates of congeners of Tegula. Data for T. funebralisare from Frank (1965) and data for T. montereyi and T. brunnea arefrom Watanabe (1982).

thermal damage will be even more of a problem forthe organism, as the limited energy available to theorganism may need to be directed away from growthand reproduction to repair.

Shifts in gene expression

In the broad context of the costs of coping withabiotic stress such as that arising from temperaturechange, the extent and nature of the shifts in gene ex-pression that are necessary for repair, replacement, andrestructuring of the cell represent an important and,now, accessible frontier of study. To date, there hasbeen no systematic study of the changes in gene ex-pression, as shown by analysis of the transcriptome(the populations of mRNA produced through tran-scriptional activity), that occur as a result of temper-ature change in any intertidal species. However, thereare data on the changes in gene expression that accom-pany the reduced access to oxygen that may occur dur-ing the tidal cycle. These data, obtained in studies ofeuryoxic intertidal and estuarine goby fish of the genusGillichthys, indicate a large-scale re-organization ofmetabolism during the shift from normoxic to hypoxicconditions (Gracey et al., 2001). In particular, genesencoding proteins involved in protein synthesis (e.g.,ribosomal proteins) and enhancement of cell prolifer-ation are strongly down-regulated in response to hyp-oxia. Conversely, genes encoding enzymes that sup-port anaerobic production of ATP and the provision ofsubstrates for these pathways (e.g., enzymes facilitat-ing conversion of amino acids into glucose units) arestrongly up-regulated. On-going studies in our labo-ratory are examining temperature-regulated shifts ofgene expression as well, to determine how extensivelygene expression is altered through the effects of thetwo environmental variables, temperature and oxygenavailability, that are apt to play especially pivotal rolesin stress during tidal cycles for intertidal species. It is,of course, premature to attempt even ‘‘back of the en-velope’’ calculations of the energetic costs of theseshifts in gene expression. Nonetheless, the costs oftranscription and translation are apt to be large and tocontribute in a measurable way to the ‘‘cost of living’’in the intertidal zone.

Growth rates of intertidal and subtidal congeners ofTegula

With the foregoing analysis of energetic costs en-tailed in living in the intertidal zone as a backdrop, Iconsider briefly the relative rates of growth of con-geners found at different vertical sites along the sub-tidal to intertidal gradient. Congeners of Tegula exhibitlarge (2–3-fold) differences in rates of growth in as-sociation with their vertical zonations (Fig. 8). T. fu-nebralis, the highest-occurring species examined, grewmost slowly; T. brunnea, was intermediate, and T.montereyi, the lowest occurring species, grew fastest(Frank, 1965; Watanabe, 1982). These data are consis-tent with, albeit they do not prove, that energetic costsare higher for species that experience the greater de-

gree of thermal stress, plus the associated costs re-sulting from longer periods of anaerobiosis and shorterperiods of feeding, found in the intertidal zone. Al-though the primary cause of the slower growth foundin the higher-occurring congeners might not be a con-sequence of energetics costs from thermal perturbationof physiological systems, these costs remain an im-portant missing piece in energy budgets and meritclose study by ecological physiologists interested invertical and latitudinal patterning of species.

SUMMARY

The widespread occurrence of adaptive variation inthe thermal sensitivities of physiological traits showsthat vertical zonation may be established and main-tained in considerable measure by a wide suite ofphysiological adaptations. The thermal sensitivities ofphysiological traits also indicate that thermal stress isapt to create significant costs to the organism, notablyin the contexts of repairing and replacing heat-dena-tured proteins and in the adaptive alterations of cellularstructures and processes during acclimatization. Thepicture that is emerging from new studies that examinetranscriptomes of environmentally stressed organismsis one in which widespread alterations in gene expres-sion and, therefore, in metabolic organization occur inresponse to environmental changes. To date, we arelikely to have observed only the tip of the iceberg ofmetabolic changes that occur in response to alterationsin the environmental factors that most affect intertidalspecies, namely temperature and access to molecularoxygen. As we more fully elucidate the nature of ther-mal stress and the responses made by organisms to thisstress, we are certain to gain important new insightsinto the interplay between physiology and vertical pat-terning along the subtidal to intertidal gradient.

ACKNOWLEDGMENTS

I thank Drs. Elizabeth Dahlhoff, Andrew Gracey,Gretchen Hofmann, Jane Lubchenco, Bruce Menge,


Eric Sanford, Jonathon Stillman, and Lars Tomanekfor stimulating discussions of many of the ideas pre-sented in this review. These studies were supported bygrants from the David and Lucile Packard Foundation(PISCO program) (this is publication #67 of the PIS-CO program) and the National Science Foundation(IBN 97-27721 and IBN 01-33184).

REFERENCES

Broekhuysen, G. J. 1940. A preliminary investigation of the impor-tance of desiccation, temperature, and salinity as factors con-trolling the vertical distribution of certain intertidal marine gas-tropods in False Bay, South Africa. Trans. Roy. Soc. S Africa28:255–292.

Connell, J. H. 1961. The influence of intra-specific competition andother factors on the distribution of the barnacle Chthamalusstellatus. Ecology 42:710–723.

Cossins, A. R. and K. Bowler. 1987. Temperature biology of ani-mals. Chapman and Hall, London.

Cossins, A. R., M. J. Friedlander, and C. L. Prosser. 1977. Corre-lations between behavioural temperature adaptations by goldfishand the viscosity and fatty acid composition of their synapticmembranes. J. Comp. Physiol. 120:109–121.

DeZwaan, A. and M. Mathieu. 1992. Cellular biochemistry and en-docrinology. In E. Gosling (ed.), The mussel Mytilus: ecology,physiology, genetics and culture, pp. 223–305. Elsevier, Am-sterdam.

Dahlhoff, E. and G. N. Somero. 1993. Effects of temperature onmitochondria from abalone (genus Haliotis): Adaptive plasticityand its limits. J. Exp. Biol. 185:151–168.

Evans, R. G. 1948. The lethal temperatures of some common Britishlittoral molluscs. J. Anim. Ecol. 17:165–173.

Feder, M. E. and G. E. Hofmann. 1999. Heat-shock proteins, mo-lecular chaperones and the stress response: Evolutionary andecological physiology. Ann. Rev. Physiol. 61:243–282.

Fields, P. A. and G. N. Somero. 1998. Hot spots in cold adaptation:Localized increases in conformational flexibility of lactate de-hydrogenase A4 orthologs of Antarctic notothenioid fishes. Proc.Natl. Acad. Sci. U.S.A. 95:11476–11481.

Frank, P. W. 1965. Shell growth in a natural population of the turbansnail, Tegula funebralis. Growth 29:395–403.

Garrity, S. D. 1984. Some adaptations of gastropods to physicalstress on a tropical rocky shore. Ecology 65:559–574.

Gowanlock, J. N. and F. R. Hayes. 1926. Contributions to the studyof marine gastropods. I. The physical factors, behavior, and in-tertidal life of Littorina. Contrib. Can. Biol. Fish. 3:133–166.

Gracey, A. Y., J. V. Troll, and G. N. Somero. 2001. Hypoxia-inducedgene expression profiling in the euryoxic fish Gillichthys mir-abilis. Proc. Natl. Acad. Sci. U.S.A. 98:1993–1998.

Hazel, J. R. 1995. Thermal adaptation in biological membranes—ishomeoviscous adaptation the explanation? Ann. Rev. Physiol.57:19–42.

Helmuth, B. S. T. 1998. Intertidal mussel microclimates: Predictingthe body temperature of a sessile invertebrate. Ecol. Monog. 68:51–74.

Helmuth, B. S. T. 1999. Thermal biology of rocky intertidal mussels:Quantifying body temperatures using climatological data. Ecol-ogy 80:15–34.

Hochachka, P. W. and G. N. Somero. 2002. Biochemical adaptation:Mechanism and process in physiological evolution. Oxford Uni-versity Press, New York.

Hofmann, G. E. and G. N. Somero. 1995. Evidence for protein dam-age at environmental temperatures: seasonal changes in levels

of ubiquitin conjugates and hsp70 in the intertidal mussel My-tilus trossulus. J. Exp. Biol. 198:1509–1518.

Hofmann, G. E. and G. N. Somero. 1996a. Interspecific variation inthermal denaturation of proteins in congeneric mussels Mytilustrossulus and M. galloprovincialis: Evidence from the heat-shock response and protein ubiquitination. Mar. Biol. 126:65–75.

Hofmann, G. E. and G. N. Somero. 1996b. Protein ubiquitinationand stress protein synthesis in Mytilus trossulus occurs duringrecovery from tidal emersion. Mol. Mar. Biol. Biotech. 5:175–184.

Logue, J. A., A. L. DeVries, E. Fodor, and A. R. Cossins. 2000.Lipid compositional correlates of temperature-adaptive differ-ences in membrane physical structure. J. Exp. Biol. 203:2105–2115.

Newell, R. C. 1979. Biology of intertidal animals. Marine EcologicalSurveys, Ltd., Faversham.

Newell, R. C. and G. M. Branch. 1980. The influence of temperatureon the maintenance of metabolic energy balance in marine in-vertebrates. Adv. Mar. Biol. 17:329–396.

Stillman, J. H. 2002. Causes and consequences of thermal tolerancelimits in rocky intertidal porcelain crabs, genus Petrolisthes.Integr. Comp. Biol. 42:000–000.

Stillman, J. H. and G. N. Somero. 1996. Adaptation to temperaturestress and aerial exposure in congeneric species of intertidalporcelain crabs (genus Petrolisthes): Correlation of physiology,biochemistry and morphology with vertical distribution. J. Exp.Biol. 199:1845–1855.

Stillman, J. H. and G. N. Somero. 2000. A comparative analysis ofthe upper thermal tolerance limits of eastern Pacific porcelaincrabs, Genus Petrolisthes: Influences of latitude, vertical zona-tion, acclimation, and phylogeny. Physiol. Biochem. Zool. 73:200–208.

Tomanek, L. 2002. The heat-shock response and patterns of verticalzonation in intertidal Tegula congeners. Integr. Comp. Biol., 42:797–807.

Tomanek, L. and G. N. Somero. 1999. Evolutionary and acclima-tion-induced variation in the heat-shock responses of congenericmarine snails (genus Tegula) from different thermal habitats:Implications for limits of thermotolerance and biogeography. J.Exp. Biol. 202:2925–2936.

Tomanek, L. and G. N. Somero. 2000. Time course and magnitudeof synthesis of heat-shock proteins in congeneric marine snails(Genus Tegula) from different tidal heights. Physiol. Biochem.Zool. 73:249–256.

Tomanek, L. and G. N. Somero. 2002. Interspecific- and acclima-tion-induced variation in levels of heat-shock proteins 70(hsp70) and 90 (hsp90) and heat-shock transcription factor-1(HSF1) in congeneric marine snails (genus Tegula): implica-tions for regulation of hsp gene expression. J. Exp. Biol. 205:677–685.

Watanabe, J. M. 1982. Aspects of community organization in a kelpforest habitat: Factors influencing the bathymetric segregationof three species of herbivorous gastropods. Ph.D. Diss., Uni-versity of California, Berkeley.

Weinstein, R. B. and G. N. Somero. 1998. Effects of temperatureon mitochondrial function in the Antarctic fish Trematomus ber-nacchii. J. Comp. Physiol. B. 168:190–196.

Williams, E. E. and G. N. Somero. 1996. Seasonal-, tidal-cycle- andmicrohabitat-related variation in membrane order of phospho-lipid vesicles from gills of the mussel Mytilus californianus. J.Exp. Biol. 199:1587–1596.

Wolcott, T. G. 1973. Physiological ecology and intertidal zonationin limpts (Acmaea). A critical look at ‘‘limiting factors.’’ Biol.Bull. 145:389–422.

somero 2002_zonacao temperatura

Documents