Sperm Chemotaxis, Fluid Shear, and the Evolution of Sexual Reproduction

Sperm chemotaxis, ﬂuid shear, and the evolution of sexual reproduction

Richard K. Zimmera,b,1,2 and Jeffrey A. Riffellc,1,2

aDepartment of Ecology and Evolutionary Biology and bNeurosciences Program and Brain Research Institute, University of California, Los Angeles, Los Angeles, CA 90095-1606; and cDepartment of Biology, University of Washington, Seattle, WA 98195

Edited by George N. Somero, Stanford University, Stanford, CA, and approved June 28, 2011 (received for review December 13, 2010) Chemical communication is fundamental to sexual reproduction, concentration gradients (11, 13) and thus promoting chemotaxis. but how sperm search for and find an egg remains enigmatic. For Previous investigations have elucidated specific mechanisms that red abalone (Haliotis rufescens), a large marine snail, the relation- underlie sperm motility (reviewed in refs. 15, 16). They have not, ship between chemical signaling and fluid motion largely deter- however, provided the information necessary for predicting emer- mines fertilization success. Egg-derived attractant plumes are gent properties of sperm–egg interactions as a consequence of dynamic, changing their size and shape in response to unique combined physical, chemical, and biological factors. combinations of physical and chemical environmental features. Soluble sperm attractants could mediate fertilization success Attractant plumes that promote sexual reproduction, however, and drive speciation, operating upstream of cell-surface proteins are limited to a precise set of hydrodynamic conditions. Perfor- and before gamete contact (17–19). Although purportedly critical mance-maximizing shears are those that most closely match flows in sexual reproduction, the influence of gamete chemical com- in native spawning habitats. Under conditions in which reproduc- munication on fertilization in flow has been inferred, but never tive success is chronically limited by sperm availability, gametes tested directly to our knowledge. Recent discoveries have con- are under selection for mechanisms that increase sperm–egg en- siderably improved knowledge on the molecular basis of signal counter. Here, chemoattraction is found to provide a cheap evolu- transduction and sperm chemotaxis (20, 21). Still, we are aware of tionary alternative for enhancing egg target size without enlarg- no experimental evidence demonstrating, unequivocally, a cause- ing cytoplasmic and/or cell volume. Because egg signaling and and-effect relationship among sperm chemoattraction, fluid mo- fi fl sperm response may be tuned to meet speci c uid-dynamic con- tion, and fertilization success. straints, shear could act as a critical selective pressure that drives The identification of a signal molecule links behavioral perfor- fi gamete evolution and determines tness. mance with ecological and evolutionary consequences. Red abalone (Haliotis rufescens) sperm detect a waterborne chemical cue invertebrate sperm chemotaxis | mammalian sperm chemotaxis | emitted by conspecific eggs, and change their swimming behavior small-scale turbulence to increase the likelihood of successful contact (22). As determined by bioassay-guided fractionation of natural egg-conditioned sea- hemical communication is pervasive among gametes of taxa water, male gamete attraction is dose-dependent and stereospe- Cwith diverse reproductive strategies. Sperm activation and cific for the L-isomer of tryptophan (23). This same compound fails chemotaxis occur in marine animals and plants that broadcast to elicit behavioral responses from the sperm of three congeneric gametes into the sea, as well as in terrestrial organisms with in- abalone species (Haliotis sorenseni, Haliotis corrugata,orHaliotis ternal fertilization (1–6). At the scale of gamete interactions (0.01– fulgens) that inhabit the same coastal environments and overlap in 1mm;Re<< 1; the Reynolds number is a dimensionless ratio, reproductive timing with H. rufescens. Waterborne compounds, reflecting the magnitude of inertial forces to viscous forces that act upon release from each of the three species, do not affect the fl on a body in ow), sperm encounter eggs while being transported sperm of red abalone. Moreover, L-tryptophan metabolites, in- within a laminar (i.e., viscous) shear flow. The magnitudes of shear cluding serotonin, tyramine, and seven other structural analogues, forces in mammalian reproductive tracts are remarkably similar to do not impact H. rufescens sperm motility or orientation toward those in many coastal ocean environments (7–14). Consequently, eggs (24). Consequently, red abalone sperm require only a single, mechanisms that drive sperm–egg interactions for external-fertil- identified compound for conspecific gamete attraction. izing marine organisms are informative of processes that operate in In this study, we simulated critical aspects of small-scale turbu- their internal-fertilizing terrestrial counterparts. lence within the natural habitats of spawning red abalone (Fig. 1). Experimental studies on sperm behavior traditionally have Laminar-shear was manipulated systematically in large Taylor– been conducted in still water. However, most gametes populate Couette flow tanks (Fig. S1). Sperm motility, gamete encounter a natural world of fluid motion. In the ocean, sperm and eggs rate, and fertilization success all were quantified, simultaneously, by usually are smaller than the tiniest vortices at these viscous- using a custom-built IR laser and computer-assisted digital imaging fl fi dominated scales (11, 12). The microscopic ow elds consist of system. Combining theoretical and empirical approaches, egg-de- — instabilities that give rise to laminar shear (10) the change in rived tryptophan plumes were described accurately and linked to velocity divided by change in distance. Such shears also are gen- sperm locomotory performance. Boundary conditions for 3D nu- erated in a terrestrial, mammalian reproductive tract, in which merical modeling (flow speed, fluid shear, tryptophan release rate flows are driven by muscular contractions, ciliary beating, and/or convective heating (8, 9, 14). Here, laminar shears result from the fl fi viscous ows produced in con ned passages. Whether gametes Author contributions: R.K.Z. and J.A.R. designed research, performed research, analyzed are suspended in a relatively unbounded (e.g., ocean) or bounded data, and wrote the paper. (e.g., vessel) medium, or attached to a surface, the magnitude of The authors declare no conflict of interest. – shear has important consequences for sperm egg interactions. This article is a PNAS Direct Submission. A given shear can positively or negatively affect gamete perfor- 1R.K.Z. and J.A.R. contributed equally to this work. mance. In excessive shears, for example, sperm may lose control of 2To whom correspondence may be addressed. E-mail: [email protected] or jriffell@ their swimming direction, which would impede chemotaxis and in- u.washington.edu. hibit fertilization (12, 13). Alternatively, higher shears could thin the This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. boundary layer in specific regions near a rotating egg, steepening 1073/pnas.1018666108/-/DCSupplemental.

13200–13205 | PNAS | August 9, 2011 | vol. 108 | no. 32 www.pnas.org/cgi/doi/10.1073/pnas.1018666108 Downloaded by guest on October 2, 2021 Fig. 1. (A) Open habitat of giant kelp forest in shallow (10–15 m depth) coastal waters offshore of Point Loma, San Diego, California (photo by Eric Hanauer). (Scale bar, 25 cm.) (B) Within this forest (and others like it), adult red abalone aggregate underneath ledges and in crevices among rocky reefs. Hydrodynamic measurements have characterized the physical properties of these environments (13). Relatively strong cross- shelf and weak longshore currents are typical of open reef habitats. In contrast, flow speeds among the rocky ledges and crevices harboring groups of red abalone (”hot spots”) are two to three times slower than in exposed areas. These hot spots exhibit significantly smaller Reynolds stresses, turbulent energy dissipation rates, and shears (ranging from 0.3 to 2.4 s−1 and − from 4.8 to 13.4 s 1 in hot spots and open habitats, respectively; ref. 13). Red abalone thus aggregate and spawn at sites where water motion is relatively tranquil. Critical aspects of these flow environments can be simulated within a Taylor– Couette apparatus (photo by Eric Hanauer). (Scale bar, 15 cm.) (C) Each adult male or female spawns gametes into the sea via excurrent tremata, small holes in the shell that connect the mantle cavity (i.e., exit site for reproductive products) and surrounding ocean. The epipodium (i.e., lateral lobe of the foot) contains many small tentacles that are used in sensing water motion; two large cephalic tentacles (not shown) pro- trude from the head (arrow at left) and function primarily in olfaction (photo by Luis Ignacio-Vilchis). (Scale bar, 1.0 cm.) (D) Spawning of sperm by a single adult male. Propulsive forces generated by the muscular contractions of its foot ultimately produce a gamete jet, or plume. In nature, both mature males and females are gravid yearround and spawn synchronously. Individuals can be systematically induced to spawn in the laboratory, and a single gravid male or female releases approximately 10 billion sperm or 3 million eggs, respectively. Profuse gamete material is therefore available at almost all times for laboratory experiments (photo by Larry Friesen and Daniel Morse). (Scale bar, 0.5 cm.)

and diffusion coefficient, egg rotational velocity) reflected actual the mixing properties of fluid into which abalone naturally spawn plumes of this natural attractant from freshly spawned abalone (13). These determinations specified the range of fluid-dynamic eggs. Accordingly, we established the basic contributions of chemi- conditions for testing in present laboratory trials. Shears were 4.8 − − cal communication and fluiddynamicstofertilizationsuccessand to 13.4 s 1 in adjacent open habitats, in contrast to 0.3 to 2.4 s 1 we identified shear as a critical selective pressure that may drive the within the native crevices and under ledges where adult red aba- EVOLUTION evolution of gamete behavior. lone live and spawn (13). Spawning abalone thus aggregated at sites where water motion was substantially retarded (Fig. 1). Results and Discussion Theoretically, surface areas and volumes of egg-derived tryp- Attractant Plumes Surrounding Eggs. Field measurements within tophan plumes peaked in still water or in weak shears and de- − giant kelp forests (Macrocystis pyrifera) previously characterized creased thereafter (Fig. 2 and Table S1). Weak shears (0.1–0.5 s 1)

Fig. 2. Theoretical concentration distributions (nmol L−1, in pseudocolor) of tryptophan surrounding red abalone eggs (black spheres) in still-water and in Taylor–Couette flows. Each plot is a 2D slice, cut through the center of an egg (A–G). White arrows denote flow velocity vectors (speeds and directions). The x axis is parallel to the direction of flow and the y axis is orthogonal to flow, but parallel to the direction of shear. Tryptophan plumes were produced through 3D numerical simulations that used a coupled convection-diffusion model, taking into account egg rotation rate at each shear, and assuming a constant and con- tinuous tryptophan release over the entire egg surface for 4 min at 0.18 fmol egg−1 min−1 (SI Materials and Methods). Model parameters (flow speed and direction, fluid shear, attractant release rate and diffusion coefficient, egg di- ameter and rotational velocity, water temperature) accurately portrayed conditions in our current experiments on sperm behavior, gamete encounter rates, and fertilization success. (Scale bar, 200 μm.)

Zimmer and Riffell PNAS | August 9, 2011 | vol. 108 | no. 32 | 13201 Downloaded by guest on October 2, 2021 and slow flows [Pe of 1.5–7.5; Peclet number is a dimensionless generated torques increasingly overwhelmed sperm behavior (13). ratio, reflecting flow speed (i.e., advection) relative to coefficient of These higher shears prevented male gametes from swimming ac- molecular diffusion for L-tryptophan) resulted in elevated con- tively across streamlines. Sperm aligned parallel to streamlines and centrations that decreased with increasing distance from an egg. cells tumbled at frequencies predicted by theory for passively The plumes thus expanded along the principal flow axis, relative to transported particles (13). Whereas weak shears promoted, strong diffusion alone (i.e., still water; Fig. 2B and Table S1). In contrast, shears inhibited sperm locomotory performance and suppressed − strong shears (2.0–10.0 s 1) and fast flows (Pe of 30–151) rapidly the attractant plume of egg-derived chemical signals. reduced tryptophan concentrations below threshold in all but a very small region near the eggs (Fig. 2 E and G). Consequently, Effects of Chemical Communication and Fluid Shear on Sperm–Egg plume-maximizing shears were those most closely simulating flows Encounter Rate and Fertilization Success. Straighter and faster paths in native spawning habitats (13). need not indicate that chemically mediated behavior increases encounter rates, or ultimately enhances fertilization success. As Effects of Chemical Communication and Fluid Shear on Sperm a function of shear, magnitudes of sperm swim speed and orientation Behavior. Results of Taylor–Couette flow experiments strongly (relative to an egg surface), male–female gamete contact rates, and supported the theoretical predictions, validating the physical model percentages of fertilized eggs, all were highly correlated (Pearson of tryptophan plume dynamics (Materials and Methods provides product–moment correlation, r2 ≥ 0.73, df = 6; P < 0.05, all com- details on Taylor–Couette apparatus and experimental protocols). parisons; Figs. 4 and 5 and Tables S2–S5). Thus, fertilization success Sperm swam faster and navigated directly toward egg surfaces could be forecasted accurately from sperm swimming behavior alone. within the predicted plumes (Fig. 3 A and B and 4 A and C). In Similar trends emerged across all sperm treatments. The per- − contrast, male gametes positioned outside of the plumes swam centages of fertilized eggs peaked at 0.1 to 0.5 s 1, and then de- slower and did not orient significantly with respect to an egg (Figs. 3 creased as shear increased. At sperm concentrations of 105 and − A and C and 4 F and H). Shear exerted a strong modulatory effect 104 cells mL 1, maximal percentages of fertilized eggs were ap- on sperm behavior (Tables S2 and S3). Swim speed and orientation proximately 1.5 times those measured in still water (Fig. 5 B and − toward an egg peaked at the weakest shears tested (0.1–0.5 s 1)and C). In contrast, the maximal value decreased (1.2 times) slightly at − − then decreased as shear strengthened. At 2.0 to 10.0 s 1, flow- a higher sperm density (106 cells mL 1), as overall fertilization

Fig. 3. (A) Representative swimming paths of individual red abalone sperm surrounding conspecific eggs in FSW as a function of shear. For each experi- − − mental treatment, a dashed line denotes the predicted behavioral threshold concentration (3 × 10 10 mol L 1). This line demarcates the theoretical active space (i.e., closer to egg) from inactive space (i.e., further from egg) of an attractant plume. Active space was defined as the portion of a plume maintaining tryptophan greater than the threshold level that caused faster sperm swimming and orientation with respect to a chemical gradient. Small circles are suc- cessive digital images of sperm heads captured at 0.033-s intervals, and each arrowhead denotes the direction of travel for an individual cell. Sperm dis- placement caused by flow was subtracted on a frame-by-frame basis, so each computer-generated path reflects the actual track swum. To eliminate selection bias, a random numbers generator was used in choosing representative paths for each flow treatment. Orientation rosettes show distributions of directional tracks by sperm populations positioned inside (B) or outside (C) the active spaces of hypothesized tryptophan plumes. For B and C, complete data sets, not representative paths, were used to establish distributions and in calculating mean unit vector lengths (r) and angular headings (θ) relative to a line between each sperm head and the center of an egg. Sperm moving directly toward an egg would follow a 0° heading. A Rayleigh test (z-value) was used to compare each mean direction against a uniform circular distribution, and to calculate the P value. Sperm orientation toward an egg was significant inside the predicted active space at 0, 0.1, 0.5, 1.0, and 2.0 s−1 (V test: u ≥ 2.65, n ≥ 26; P < 0.04, all comparisons).

13202 | www.pnas.org/cgi/doi/10.1073/pnas.1018666108 Zimmer and Riffell Downloaded by guest on October 2, 2021 Fig. 4. Effects of fluid shear on direction of sperm swimming with respect to an egg (A and F) (as described in more detail in Fig. 3), direction of sperm swimming with respect to flow (B and G), sperm translational swim speed (C and H), sperm encounter rate with a theoretical, tryptophan active space surrounding an egg (D), and sperm–egg encounter rate (E). Male gametes were imaged while swimming either “inside” or “outside” the active spaces of theoretical tryptophan plumes. Complete data sets, not representative paths, were used to establish mean unit vector lengths (r) with respect to an origin in A, B, F, and G. Experiments were performed in the presence of FSW alone (solid line) or with addition of active or denatured tryptophanase (dotted or dashed lines, respectively). Each dependent variable was described as a function of log-shear, using least-squares regression to establish the best fit(F tests for FSW and denatured enzyme treatments: F ≥ 7.39, df ≥ 1, 116; P < 0.001, all comparisons). Symbols are mean values (±SEM), and error bars are smaller than symbol sizes in some cases.

levels approached saturation (an asymptote of 100% eggs fertil- (Fig. 5 and Table S5). Weak shears therefore promoted sperm ized; Fig. 5A). Compared with still water, fertilization success was chemoattraction as well as reproductive success. − elevated significantly at 0.1 to 0.5 s 1, was approximately the same As always, correlation does not imply causation. Whereas results − − at 1.0 to 2.0 s 1, and was depressed significantly at 4.0 to 10.0 s 1 showed a strong association between egg-derived attractant plumes and sperm behavioral performance, these experiments were not designed to show a cause-and-effect relationship. Con- sequently, we determined whether eliminating the chemical signal around eggs would prevent fertilization. Freshly spawned eggs and sperm were placed in Taylor–Couette chambers containing filtered seawater (FSW) as before, but now with addition of activated or

− EVOLUTION denatured (i.e., boiled) tryptophanase (2 μgmL 1). This enzyme, when active, selectively digests free tryptophan in solution. The addition of activated tryptophanase had profound consequences for sperm–egg interactions. First, the enzyme did not affect sperm membranes, receptors, or behaviors, or the proclivity of male or female gametes for fertilization (22). It did, however, ex- tinguish the signal surrounding an egg, as evidenced by sperm in- ability to navigate within hypothesized plumes, even from a distance of 100 μm, or less (Fig. 6 and Tables S6 and S7). HPLC indicated no measurable accumulation of tryptophan in seawater, when both enzyme and eggs were present. Second, elimination of the tryptophan and sperm chemoattraction precipitated a significant decrease in gamete encounter rate and fertilization success (Figs. 4 and 5 and Tables S8 and S9). Conversely, there was no decay in sperm navi- gation (toward an egg) and swim speed, encounter rate, or fertilization with the denatured tryptophanase (Figs. 4 and 5, Fig. S2,and Tables S6–S9). Tryptophan release by eggs, therefore, was a causa- tive agent and critical determinant of fertilization success. Sperm chemoattraction and fluid shear each had significant effects on fertilization dynamics. Which process plays the ascen- dant role? To answer this question, we performed a series of stepwise multiple regressions on the fertilization data. Taken as a whole, our experiments measured percentages of fertilized eggs − over a wide range of shears (0–10 s 1, from abalone spawning to − open kelp forest habitats), sperm densities (104–106 cells mL 1, Fig. 5. Effects of fluid shear on fertilization success (i.e., percentage of from sperm-limiting to sperm-saturating conditions), and in the − fertilized eggs). Egg density was held constant (103 cells mL 1) and, in sep- presence of active or denatured tryptophanase. Shear—not che- 6 5 4 arate treatments, sperm density was tested at (A)10 ,(B)10 ,or(C)10 cells moattraction—explained most (55–64%) of the variation in fer- −1 − mL . Experiments were performed in the presence of FSW alone (solid line) tilization success at low sperm densities (104 and 105 sperm mL 1; or with addition of active or denatured tryptophanase (dotted or dashed − Table S10). In contrast, at a high sperm density (106 mL 1), che- lines, respectively). Fertilization success was described as a function of log- fi shear, using least-squares regression to establish the best fit(F tests: F ≥ 50.5, moattraction had a signi cantly greater impact (67% of variation df ≥ 1, 58; P < 0.001, all comparisons). Symbols are mean values (±SEM), and in fertilization). Thus, shear dominated chemical communication error bars are smaller than symbols in some cases. only under limiting sperm conditions. Shear did not damage either

Zimmer and Riffell PNAS | August 9, 2011 | vol. 108 | no. 32 | 13203 Downloaded by guest on October 2, 2021 Fig. 6. Effects of ﬂuid shear and tryptophanase on representative swimming paths of individual red abalone sperm (A) and on orientation distributions of directional tracks by sperm populations positioned inside (B) or outside (C) of theoretical tryptophan plumes surrounding eggs. All experimental procedures and analyses, except for enzyme addition, were the same as described for Fig. 3. A Rayleigh test (z-value) was used to compare each mean direction against a uniform circular distribution, and to calculate the P value. Sperm orientation toward an egg was not signiﬁcant in still water and at each tested shear (V tests: u ≤ 1.38; P ≥ 0.39).

sperm or eggs (13). Instead, acting to facilitate or inhibit, it mod- a rich biochemical environment for synthesizing natural products ulated the strength of chemically mediated, gamete interactions. after fusion, as required in embryo development (31). In contrast, the free amino acid L-tryptophan is taken up by maternal abalone Chemical Communication, Fluid Shear, and Evolution of Sexual from a dietary source and incorporated directly in egg cytoplasm Reproduction. The evolution of gamete size and morphology is during oogenesis. Thus, signal production, as well as release (via a major unsolved problem in reproductive biology. Under con- diffusion), consumes little or no metabolic energy and expends less ditions in which reproductive success is chronically limited by than 1% of total cytoplasmic tryptophan reserves (24). Whereas sperm availability, adults and gametes are under selection for tryptophan acts as a sperm attractant, it also is a precursor for mechanisms that increase sperm–egg contact (25). One such synthesizing many neurotransmitters and neuromodulators. As mechanism could involve changes in the physical size of the egg, a metabolic substrate essential to the development of the larval because enlarging the “target” increases the probability of sperm– nervous system, tryptophan could be an honest indicator of egg egg collision (25–27). Models of evolution have focused, tradi- fitness for prospective sperm suitors (32). Furthermore, red abalone tionally, on postzygotic consequences of egg size for larval or ju- eggs stop releasing tryptophan as they age and become infertile venile survivorship (28, 29). Another implication of the target size (24). Our results, therefore, suggest that endogenous signaling hypothesis, however, is that prezygotic benefits to fertilization pathways have been coopted for external communication, as an could drive the evolution of egg size and, in turn, anisogamy (25). adaptation to increase the likelihood of reproductive success. Be- To date, theoretical models of gamete size evolution have not cause egg signaling and sperm response are possibly tuned to meet considered effects of fluid motion on sperm–egg encounter specific fluid-dynamic constraints, shear may act as a critical selec- probabilities. Because shear is a natural feature of nearly all re- tive pressure that drives gamete evolution and determines fitness. productive habitats, it may exert strong selective pressure on gamete morphology. In fact, shear initiates egg rotation at an Materials and Methods angular velocity directly proportional to gamete size (i.e., radius) Procedures for the maintenance and spawning of abalones, instrumentation, (11, 13, 30). As a consequence of this rotation, fluid accelerates analysis and construction of the Taylor–Couette apparatus, chemosensory when it approaches an egg, compressing or closing streamlines treatments on sperm behavior, and theoretical flow fields and attractant and locally increasing shear stress near an egg surface. The like- plumes around eggs are detailed in SI Materials and Methods. lihood of sperm “slipping” around an egg surface, rather than encountering it, increases significantly with rotation rate (13). Effects of Chemical Communication and Fluid Shear on Fertilization Success. Relationships among fluid shear, chemical communication, and fertilization Thus, a larger egg is not always a better target. success were determined in a Taylor–Couette apparatus. This device consists For red abalone, chemoattraction provides a cheap evolutionary of two nested cylinders with radii of 6.1 cm and 6.9 cm for the inner and alternative for increasing egg target size without enlarging cyto- outer cylinders, respectively (Fig. S1; details of theory and construction are plasmic and/or cell volume. Egg cytoplasm is an expensive com- provided in refs. 11, 12). The gap between the cylinders was filled with modity. It contains a vast array of organic molecules and provides sperm at a concentration of 104,105,or106 cells mL−1. In each trial, the gap

13204 | www.pnas.org/cgi/doi/10.1073/pnas.1018666108 Zimmer and Riffell Downloaded by guest on October 2, 2021 − contained FSW alone, FSW with tryptophanase (2 μgmL1), or FSW with substituted for their live counterparts, and the measurements repeated (n =8 − denatured (i.e., boiled) tryptophanase (2 μgmL 1). Before use, tryptopha- replicate trials for each shear). The dead sperm served as passive particles, − nase was activated by incubation with 100 μmol L 1 pyridoxal-5′-phosphate revealing background fluid movement experienced by live gametes. From in FSW (pH 7.9) at 37 °C for 1 h to ensure maximum formation of the paths of dead sperm, 2D velocity fields were mapped with respect to positions holoenzyme (22). The addition of denatured tryptophanase controlled for (x,y coordinates) within the gap. The computed velocities based on dead nonspecific effects of increased protein concentration. A computer-con- sperm paths were then subtracted from live sperm paths on a frame-by-frame trolled stepper motor system was activated after all visible air bubbles were basis, by using a customized MATLAB program (as detailed in ref. 13). purged from the tank (standing upright). Within 5 s, this unit brought the Elimination of the flow component from a live cell path revealed the −1 spinning cylinders to a designated shear of 0.1, 0.5, 1.0, 2.0, 4.0, or 10.0 s . direction of swimming within a circular coordinate system. Here, θ is the For comparison, trials also were performed in still water, without motor angle relative to an operationally defined origin (0°) and r is the unit vector activation. Eggs were placed in custom-built mesh tubes in the absence of for n cells in a population: flow. Because filament size was inconsequential compared with the size of fi X X 1=2 the mesh openings, these compartments failed to signi cantly affect sperm 1 2 2 r ¼ cos θ þ sin θ [1] swim speeds or directions under test conditions (13). Ten to 15 replicate trials n i i were conducted for each sperm concentration, shear (or still water), and enzyme treatment (N = 773 total trials). A unit vector length of 1 indicates that all sperm swam on a single trajectory Shear effects on fertilization were quantified at a single contact time. for a given treatment. In contrast, a length of zero denotes random move- Fifteen seconds after egg introduction, 10 mL of mixed gamete suspension ments without a shared bearing. Two relative orientations were evaluated: werewithdrawn from the middleof the gap, 4 cm below the water surface. This did cells orient relative to flow or egg surface? If the former, cells swimming in time reflected a short, but realistic, gamete encounter interval in field envi- straight paths, downstream or upstream, would move with an θ of 0° or 180°, ronments (33, 34). Moreover, we previously calculated the Kolmogorov time respectively. The coordinate system was set up with the origin facing away scales from hydrodynamic measurements of flows within the crevices and from flow. According to the latter, sperm swimming trajectory was evalu- ledges that were inhabited by adult red abalone (13). These scales reflect the ated with respect to the nearest egg. In this case, 0° was defined as the angle lifetimes of the tiniest turbulent eddies (called the Kolmogorov microscales) in between each cell body and the center of the egg. Sperm moving directly which spawned sperm and eggs resided naturally, before kinetic energy dis- toward an egg thus would follow a 0° heading. For each treatment, these sipation by fluid viscosity to heat. As determined empirically, Kolmogorov two separate analyses were performed on the same data set. A Rayleigh test times of 2 to 13 s compared favorably to the duration of our current fertil- was applied initially to compare the mean direction swum against a uniform ization assays and thus provided an immediate ecological context (13). In circular distribution. If significantly different, a V test was used to determine ∼ 3 −1 present experiments, eggs (at 10 mL ) were captured from suspension on the fit with respect to each of the origins (0°) as defined earlier. a 100-μm mesh screen, and then rinsed thoroughly with 50 mL of FSW. Re- peated microscopic examinations indicated that the rinse eliminated all sperm ACKNOWLEDGMENTS. This paper is dedicated to the memory of Mia J. from egg surfaces, except those attached to the vitelline envelope. After 3-h Tegner, a leading researcher for the preservation of endangered abalone incubation in FSW, eggs were fixed in 5% buffered formalin and assessed for species, who provided resources, lab space for the performed experiments, percentage fertilized as cleavage into four- or eight-celled embryos. as well as valuable advice during this project. We thank Kristin L. Riser and L. Ignacio Vilchis for assistance with field and laboratory tasks. Paul K. Dayton Measurements of Sperm Swimming Speed and Direction in Shear Flows. Timing and Michael I. Latz provided laboratory facilities and Cheryl Ann Zimmer helped us conceptualize the physics and put our thoughts into words. This and precise location of each encounter between sperm and egg was recorded research was supported by National Science Foundation Award IOS 08-20645 from digital images. Swim speed and position of each sperm cell also were and DBI 11-21692, National Oceanic and Atmospheric Administration

× μ fi × fl EVOLUTION determined within a 1,000 800 m eld (parallel perpendicular to ow California Sea Grant College Program Project R/F-197, National Institutes direction) surrounding an egg. To control for effects of flow on measure- of Health Grant 2-K12-GM000708, and the University of California, Los ments of swim speed, heat-killed sperm (20 min exposure at 40 °C) were Angeles, Council on Research.

1. Gakamsky A, Armon L, Eisenbach M (2009) Behavioral response of human sperma- 20. Carlson AE, Hille B, Babcock DF (2007) External Ca2+ acts upstream of adenylyl cyclase tozoa to a concentration jump of chemoattractants or intracellular cyclic nucleotides. SACY in the bicarbonate signaled activation of sperm motility. Dev Biol 312:183–192. Hum Reprod 24:1152–1163. 21. Wood CD, et al. (2007) Altering the speract-induced ion permeability changes that 2+ 2. Maier I, Muller DG (1986) Sexual pheromones in algae. Biol Bull 170:145–175. generate flagellar Ca spikes regulates their kinetics and sea urchin sperm motility. 3. Miller RL (1985) Demonstration of sperm chemotaxis in Echinodermata: Asteroidea, Dev Biol 306:525–537. Holothuroidea, Ophiuroidea. J Exp Zool 234:383–414. 22. Riffell JA, Krug PJ, Zimmer RK (2004) The ecological and evolutionary consequences 4. Olson JH, et al. (2001) Allurin, a 21-kDa sperm chemoattractant from Xenopus egg jelly, is of sperm chemoattraction. Proc Natl Acad Sci USA 101:4501–4506. related to mammalian sperm-binding proteins. Proc Natl Acad Sci USA 98:11205–11210. 23. Riffell JA, Krug PJ, Zimmer RK (2002) Fertilization in the sea: The chemical identity of – 5. Spehr M, et al. (2003) Identification of a testicular odorant receptor mediating human an abalone sperm attractant. J Exp Biol 205:1439 1450. sperm chemotaxis. Science 299:2054–2058. 24. Krug PJ, Riffell JA, Zimmer RK (2009) Endogenous signaling pathways and chemical – 6. Ward GE, Brokaw CJ, Garbers DL, Vacquier VD (1985) Chemotaxis of Arbacia punctulata communication between sperm and egg. J Exp Biol 212:1092 1100. 25. Podolsky RD (2004) Life-history consequences of investment in free-spawned eggs spermatozoa to resact, a peptide from the egg jelly layer. JCellBiol101:2324–2329. – 7. Denny MW, Gaylord B (2010) Marine ecomechanics. Annu Rev Mar Sci 2:89–114. and their accessory coats. Am Nat 163:735 753. 26. Dusenbery DB (2006) Selection for high gamete encounter rates explains the evolu- 8. Eytan O, Jaffa AJ, Elad D (2001) Peristaltic flow in a tapered channel: Application to tion of anisogamy using plausible assumptions about size relationships of swimming embryo transport within the uterine cavity. Med Eng Phys 23:473–482. speed and duration. J Theor Biol 241:33–38. 9. Fauci LJ, Dillon R (2006) Biofluidmechanics of reproduction. Annu Rev Fluid Mech 38:371–394. 27. Levitan DR (1993) The importance of sperm limitation to the evolution of egg size in 10. Jumars PA, Trowbridge JH, Boss E, Karp-Boss L (2009) Turbulence-plankton inter- marine invertebrates. Am Nat 141:517–536. actions: A new cartoon. Mar Ecol (Berl) 30:133–150. 28. Randerson JP, Hurst LD (2001) The uncertain evolution of the sexes. Trends Ecol Evol 11. Karp-Boss L, Boss E, Jumars PA (1996) Nutrient fluxes to planktonic osmotrophs in the 16:571–579. presence of fluid motion. Oceanogr Mar Biol Annu Rev 34:71–107. 29. Vance RR (1973) On reproductive strategies in marine benthic invertebrates. Am Nat 12. Mead KS, Denny MW (1995) The effects of hydrodynamic shear stress on fertilization 107:339–352. and early development of the purple sea urchin Strongylocentrotus purpuratus. Biol 30. Jeffery GB (1922) The motion of ellipsoidal particles immersed in a viscous fluid. Roy – Bull 188:46 56. Soc Lond A 102:161–179. fl fl 13. Riffell JA, Zimmer RK (2007) Sex and ow: The consequences of uid shear for sperm- 31. Jaeckle WB (1995) Variation in the size, energy content, and biochemical composition – egg interactions. J Exp Biol 210:3644 3660. of invertebrate eggs: correlates to the mode of larval development. Ecology of 14. Winet H, Bernstein GS, Head J (1984) Observations on the response of human sper- Marine Invertebrate Larvae, ed McEdward LR (CRC Press, Boca Raton), pp 49–77. matozoa to gravity, boundaries and fluid shear. J Reprod Fertil 70:511–523. 32. Byrne M, Sewell MA, Selvakumaraswamy P, Prowse TA (2006) The larval apical organ 15. Eisenbach M, Giojalas LC (2006) Sperm guidance in mammals - an unpaved road to in the holothuroid Chiridota gigas (Apodida): inferences on evolution of the Ambu- the egg. Nat Rev Mol Cell Biol 7:276–285. lacrarian larval nervous system. Biol Bull 211:95–100. 16. Kaupp UB, Kashikar ND, Weyand I (2008) Mechanisms of sperm chemotaxis. Annu Rev 33. Levitan DR (1998) Does Bateman’s principle apply to broadcast-spawning organisms? Physiol 70:93–117. Egg traits influence in situ fertilization rates among congeneric sea urchins. Evolution 17. Vacquier VD (1998) Evolution of gamete recognition proteins. Science 281:1995–1998. 52:1043–1056. 18. Ward CR, Kopf GS (1993) Molecular events mediating sperm activation. Dev Biol 158:9–34. 34. Pennington JT (1985) The ecology of fertilization of echinoid eggs: the consequences 19. Zimmer RK, Butman CA (2000) Chemical signaling processes in the marine environ- of sperm dilution, adult aggregation, and and synchronous spawning. Biol Bull 169: ment. Biol Bull 198:168–187. 417–430.

Zimmer and Riffell PNAS | August 9, 2011 | vol. 108 | no. 32 | 13205 Downloaded by guest on October 2, 2021