Caribbean spiny lobsters equally avoid dead and clinically PaV1-infected conspecifics

Abstract

Social behaviour in Caribbean spiny lobsters (Panulirus argus) is mediated by conspecific chemical cues. These lobsters can be attracted to shelters emanating chemical cues from conspecifics but tend to avoid shelters emanating chemical cues from injured conspecifics, dead conspecifics, and conspecifics with visible signs of a potentially lethal disease caused by the pathogenic Panulirus argus virus 1 (PaV1). However, previous studies have not controlled for the presence of PaV1 (i.e. subclinical infection) in grossly "healthy" lobsters, although visible signs of disease do not appear until several weeks after infection. We conducted a controlled experiment using a set of 2 m-long Y-mazes to examine and contrast the response of P. argus lobsters to shelters emanating chemical cues from conspecifics in four different conditions: uninfected, subclinically PaV1-infected (i.e. infected but not diseased), clinically PaV1-infected (i.e. infected and diseased), and dead. Using polymerase chain reaction, we tested for PaV1 in all grossly healthy lobsters and used exclusively uninfected lobsters in intermolt as focal lobsters. Focal lobsters similarly avoided shelters emanating chemical cues from clinically infected (80% avoidance) and from dead conspecifics (85% avoidance), but their response to chemical cues from uninfected and from subclinically infected conspecifics did not differ significantly from random. These results indicate that PaV1-diseased lobsters produce chemical cues that are as repellent to conspecifics as are chemicals emanating from dead conspecifics, and that subclinically infected lobsters either do not emit the repellent chemicals or they do so at sub-threshold levels. However, the nature of the repellent chemicals and whether they originate from the pathogen or the host remains to be determined.