Responses to damage in an arborescent bryozoan: Effects of injury location

https://doi.org/10.1016/j.jembe.2005.04.008Get rights and content

Abstract

A common consequence of the predation or physical disturbance of sessile modular animals is the removal of part of the colony. This damage is often sub-lethal, and leaves the remainder of the colony to repair and regenerate lost tissue. Damage may, however, have lasting effects on the life history characteristics of a colony. We examined the effects of damage on growth and reproduction in the arborescent bryozoan Bugula neritina (Linnaeus), with a particular focus on the effects of the location of that damage. We predicted that a damaged colony would initially allocate more energy to damage repair and subsequent growth, increasing in size to decrease the risk of mortality, at the expense of its short-term reproductive potential. A short-term experiment assessed the immediate regenerative capacities of colonies, while a longer-term experiment measured overall colony responses to damage at different colony locations. In this experiment, zooids were removed in three colony regions; all on one colony branch, spread evenly across all growing tips, and a third, intermediate pattern. A third experiment tested the effects of predator-induced damage versus damage alone, by measuring the recovery of colonies subjected to either a predation event by the nudibranch Polycera hedgpethi, or to an imitation predation event, where an equivalent section of the colony was removed manually. Regeneration was rapid, with budding resuming a few days after damage occurred. Removing around 20% of the colony did not consistently alter colony growth rates in any experiment, but the onset of reproduction was delayed by 7 to 20 days, and subsequent colony fecundity reduced by as much as 70% in one damage treatment. The removal of the same amount of tissue by nudibranch predators, or manually removed in a treatment that mimicked this predation, had no detrimental effect on the onset of reproduction or the eventual reproductive output. Nudibranchs removed tissue by consuming whole colony branches, and the experimental treatment that removed one branch also had the least effect of three damage treatments. These results suggest the severity of long-term effects depends on the location of the injury, and that B. neritina does not modify its growth or reproductive patterns to minimise these effects.

Introduction

Disturbance, through wave action, storms, predation and human interference, is a ubiquitous feature of marine ecosystems, and may act to modify systems by changing overall species composition, or moulding the life histories of individual organisms (e.g. Hunter, 1993, Connell and Keough, 1985, Dial and Roughgarden, 1998, Keough and Quinn, 1998). The ability of a species to persist in a habitat with frequent disturbance is largely determined by the ways in which individual organisms respond. If primary defence mechanisms cannot prevent damage through disturbance, then the ability to recover becomes crucial, and the life history characteristics of an animal determine that recovery response.

Modular animals perform differently to unitary animals after disturbance (Jerling, 1985, Vuorisalo and Tuomi, 1986). Disturbances may remove whole unitary animals, but often only a part of a modular animal is destroyed. Because no single module is essential to colony survival, regeneration and recovery are possible (e.g. Highsmith, 1982, Lasker, 1984, Karlson, 1986, Karlson, 1988, Chadwick and Loya, 1990, Smith and Hughes, 1999). Colonies may also undergo a number of changes that complicate their demography, such as fission, forming smaller colonies, or fusing with compatible neighbours to form a larger colony comprising modules of varying ages and genotypes. Throughout a colony's lifetime, it may go through each of these processes several times. As a result, there is often a decoupling of age and size in modular animals, and age becomes not only difficult to determine, but also a poor predictor of reproductive capacity (e.g. Hughes and Jackson, 1980). Understanding the population dynamics of modular animals relies in part on understanding the processes that damage colonies, and the nature of colony responses to these processes.

Conventional life history theory established for unitary organisms does not account for the growth and reproductive patterns of modular organisms, since the demography of unitary organisms is based on age, yet complex patterns of damage and recovery in modular animals mean that age and size are not always linked (Hughes and Jackson, 1980, Stearns, 1992, Hall and Hughes, 1996). Initially, it was thought that size-based demographic models would be appropriate for modular animals (e.g. Hughes, 1984), with an increase in colony size through the addition of new modules generally indicating a higher reproductive capacity and lower likelihood of whole-colony mortality. More recent conceptual models, however, invoke combined effects of colony size and age (Hughes and Connell, 1987), and consider effects of the age structure of modules within a colony (Jackson and Hughes, 1985). The damage history of a colony can alter the ages of its modules, interacting with the effects of total colony size to determine growth and reproductive capabilities. Different kinds of damage can also affect colony responses. For example, Hall (2001) assessed the effects of tissue loss, scraping injury and branch removal on a number of branching, massive and submassive coral species, finding large differences in response between species, but also after different injury types. These differences may reflect the relative ease with which colonies can recover space after being damaged. Tissue removal alone is less destructive in the long term than the removal of a branch, since the former leaves the skeleton intact. We may thus expect responses to damage to be mediated by the effects of that damage on a colony's ability to feed, grow and recover space.

In modular organisms, growth by the asexual propagation of modules and the (usually) sexual production of dispersive propagules are continually in competition for finite resources (Sebens, 1979, Sebens, 1987, Sebens and Thorne, 1985). Repairing damage in mature colonies entails the reallocation of resources from propagules to repair and the production of new modules. A colony's ability to transport these resources can be affected by its growth form and the level of integration of its modules. There is growing understanding of the specific mechanisms of resource allocation in modular animals (Bobin, 1977, Lutaud, 1983, Harper, 1985, Mukai et al., 1997, Oren et al., 1997, Best and Thorpe, 1985, Best and Thorpe, 2002), and the way a colony responds after damage is also affected by the efficiency of resource transfer between colony regions. A colony-wide response may signify a high capacity for the translocation of resources between regions, while a localised response may mean that only local energy stores are being used in the repair process (Meesters et al., 1994, Ruohomäki et al., 1997). This response should depend on the severity of the injury, with local resources being utilised in the repair of relatively minor damage, and more extensive damage requiring additional resources from other colony regions. It is also possible that these different types of responses can be seen within a single colony. For example, in the encrusting bryozoan Membranipora membranacea, obstructing colony growth on one side resulted in a higher relative growth rate on the unobstructed side, presumably a compensatory response, while damage induced reproduction in zooids adjacent to the damage site (Harvell and Helling, 1993). We suggest that differences in the severity of a single type of damage would influence the magnitude of the colony's response, whereas changing the pattern of damage while keeping the severity constant might influence the nature and location of the response. After such different patterns of damage, the effects across a colony should vary widely if responses are localised, but be relatively consistent if modules within a colony are well integrated.

Most of our understanding of the effects of damage has come from studies of encrusting and sheet-like cnidarians, bryozoans and ascidians (e.g. Bak et al., 1981, Palumbi and Jackson, 1982, Hughes, 1984, Karlson, 1986, Karlson, 1988, Harvell and Helling, 1993, Klemke, 1993, Meesters et al., 1994), with studies of other growth forms coming predominantly from tropical cnidarians (Wahle, 1985, Lasker, 1984, Hall, 2001). Here we describe responses to damage in a temperate arborescent bryozoan, Bugula neritina. We describe basic colony responses (production of modules, onset of reproduction, and fecundity) to damage, and then compare responses to variation in the location of module loss, recording responses averaged across a whole colony, as well as responses in different colony regions. The efficiency of the recovery response may reliably denote the extent of nutrient transfer between colony regions, since a damaged region cannot recover effectively without nutrient input from adjacent feeding zooids. Specifically, if there is insufficient nutrient transfer between colony regions to allow adequate recovery of the damaged areas, control colonies and damaged colonies should differ in their colony-wide investment in growth and reproduction. Alternatively, we could expect a well-integrated colony to overcompensate for damage to a particular region by investing energy into continued growth of the undamaged colony area, where resource pathways are uninterrupted. In this case, overall growth in a damaged colony would be less than or no different to that of the controls, but growth of the undamaged colony side would be greater than that of control colonies.

We produced most damage by excising parts of the colony, but in a later experiment, we assessed the effects of one important agent of damage, the predatory nudibranch Polycera hedgpethi. We compared the effects from damage caused by the nudibranch to those of excising an equivalent section of the colony manually, to assess whether the presence of the predator affects the recovery response.

Section snippets

Materials and methods

The arborescent cheilostome bryozoan Bugula neritina (Linnaeus) is a species complex originating on the west coast of the USA. Analysis of mitochondrial COI sequences from B. neritina colonies collected along the southern Australian coastline, including the study area of northern Port Phillip Bay, has revealed only the type S “shallow” strain of Davidson and Haygood (1999) (Mackie, 2002). Throughout this paper, we refer to the S variant of B. neritina. The abundance of B. neritina colonies in

Short-term colony regeneration

All colonies began to recover quickly after damage. Damaged colony sides were smaller than undamaged sides across the entire experimental period, and at the final census, undamaged colony sides were larger than the damaged sides (Fig. 2a), although this result was not significant (df = 20, t = 2.030, p = 0.056). Damaged sides exhibited slower initial growth rates, but after 4 days, growth profiles were fairly similar. Undamaged sides of colonies showed budding on a high proportion of the branch tips,

Discussion

Damaging B. neritina colonies before they started to reproduce had lasting negative effects on growth rates, reproductive onset and subsequent reproductive performance. Importantly, the location of the inflicted damage was the most influential factor governing the long-term reproductive potential of colonies. Colonies that suffered greater losses of the growing tissue at the tips of branches delayed reproduction to a larger extent than colonies where damage was concentrated along a single

Acknowledgments

This work was supported by a grant from the Australian Research Council to MJK. [AU]

References (50)

  • S.K. Davidson et al.

    Identification of sibling species of the bryozoan Bugula neritina that produce different anticancer bryostatins and harbor distinct strains of the bacterial symbiont “Candidatus Endobugula sertula”

    Biol. Bull.

    (1999)
  • R. Dial et al.

    Theory of marine communities: the intermediate disturbance hypothesis

    Ecology

    (1998)
  • J.B.C. Jackson et al.

    Adaptive strategies of coral-reef invertebrates

    Am. Sci.

    (1985)
  • L. Jerling

    Are plants and animals alike? A note on evolutionary plant population ecology

    Oikos

    (1985)
  • V.R. Hall et al.

    Reproductive strategies of modular organisms: comparative studies of reef-building corals

    Ecology

    (1996)
  • J.L. Harper

    Modules, branches and the capture of resources

  • C.D. Harvell

    Predator-induced defense in a marine bryozoan

    Science

    (1984)
  • C.D. Harvell

    Why nudibranchs are partial predators: intracolonial variation in bryozoan palatability

    Ecology

    (1984)
  • C.D. Harvell

    Coloniality and inducible polymorphism

    Am. Nat.

    (1991)
  • C.D. Harvell et al.

    Experimental induction of localized reproduction in a marine bryozoan

    Biol. Bull.

    (1993)
  • M. Heino et al.

    Evolution of resource allocation between growth and reproduction in animals with indeterminate growth

    J. Evol. Biol.

    (1999)
  • R.C. Highsmith

    Reproduction by fragmentation in corals

    Mar. Ecol. Prog. Ser.

    (1982)
  • T.P. Hughes

    Population dynamics based on individual size rather than age: a general model with a coral reef example

    Am. Nat.

    (1984)
  • T.P. Hughes et al.

    Population dynamics based on size or age? A reef-coral analysis

    Am. Nat.

    (1987)
  • T.P. Hughes et al.

    Multiple stressors on coral reefs: a long-term perspective

    Limnol. Oceanogr.

    (1999)
  • Cited by (0)

    View full text