Photobacterium damselae subsp. damselae, a bacterium pathogenic for marine animals and humans

1Photobacterium damselae subsp. damselae is a marine bacterium of the family Vibrionaceae that causes infections in a variety of marine animals and also in humans. A bit of historic perspective is necessary in order to understand its current taxonomic placement as well as the changes in its nomenclature during the past decades. In 1971 an “unnamed marine Vibrio” was isolated as the causative agent of a human infectious case. Later, this same organism was isolated from skin ulcers of damselfish (Chromis punctipinnis) and the name Vibrio damsela was first coined. Further genetic and phenotypic studies indicated that the strains of V. damsela were closely related to species of the genus Photobacterium, and the name Photobacterium damsela was proposed. In 1995, DNA–DNA hybridization data and 16S rRNA sequence analysis demonstrated that Photobacterium damsela was closely related to a fish pathogen formerly named Pasteurella piscicida, the causative agent of pasteurellosis in fish. Hence, these two organisms were assigned to the same species epithet, Photobacterium damselae, with category of subspecies, Photobacterium damselae subsp. damselae and Photobacterium damselae subsp. piscicida respectively. Despite their similarity at the 16S gene sequence and the high percentage of DNA–DNA relatedness between them, these two subspecies are clearly distinguished by several phenotypical traits. Differential phenotypical tests of interest for subspecies discrimination that are positive only for subsp. damselae include motility, nitrate reduction and hemolysis on sheep blood agar. Of special relevance is the ability of most subsp. damselae strains to grow at 37°C (a temperature inhibitory for subsp. piscicida), a trait that allows Photobacterium damselae subsp. damselae to potentially colonize and establish an infection in a homeotherm animal.

2Photobacterium damselae subsp. damselae is an autochthonous member of aquatic ecosystems. Strains of this pathogen have been isolated from sea and estuarine waters, from seaweeds, from apparently uninfected marine animals and from seafood, and it is considered a common member of the natural microbiota of healthy carcharhinid sharks.

3In addition, Photobacterium damselae subsp. damselae is considered a primary pathogen of several species of wild fish (damselfish, catfish, shark, stingray, etc.), as well as of fish species of economical importance in aquaculture, causing wound infections and hemorrhagic septicemia. Cultivated species reported to be affected by this pathogen include turbot (Psetta maxima; ), rainbow trout (Oncorhynchus mykiss; ), ovate pompano (Trachinotus ovatus; ), eel (Anguilla reinhardtii; ), sea bream (Sparus aurata; ), sea bass (Dicentrarchus labrax), yellowtail (Seriola quinqueradiata), redbanded seabream (Pagrus auriga), white seabream (Diplodus sargus), and meagre (Argyrosomus regius;,, ), among others. The recent first reports on isolation of this pathogen from diseased marine fish of new cultured species, suggest that Photobacterium damselae subsp. damselae can be considered as an emerging pathogen in marine aquaculture.

4Moreover, Photobacterium damselae subsp. damselae has been isolated as a pathogen of brown shark (Carcharhinus plumbeus; ), of reptiles as the leatherback sea turtle (Dermochelys coriacea; ), molluscs (Octopus joubini; ), crustaceans, dolphins (Tursiops truncatus and Delphinus delphis;; ) and Bryde’s whale (Balaenoptera edeni; ).

5Virulent isolates are capable of survival in seawater microcosms at 14–22°C as culturable bacteria for long periods of time, maintaining their infectivity for fish. Similarly, this pathogen can infect new fish hosts through water, and the spread of the disease depends largely on water temperature and salinity. Typical signs of the disease in infected fish include hemorrhaged areas on the body surface and ulcerative lesions. In damselfish, ulcers typically occur in the region of the pectoral fin and caudal peduncle and may reach 5–20 mm in diameter, while in turbot the most remarkable symptoms are extensive hemorrhages in eyes, mouth, and jaws.

6Experimental inoculation of Photobacterium damselae subsp. damselae extracellular products (ECPs) in a redbanded seabream model was reported to cause lethargy, increase in the respiratory frequency, mucus production, presence of ascitic liquid, hemorrhagic and enlarged liver, and hemorrhages in the abdominal cavity. A histological analysis of internal organs in experimentally infected turbot indicated that the ECPs and cells of virulent strains cause similar tissue damage. Structural changes included destruction and necrosis of cells, as well as accumulation of blood cells in interstitial tissue.

7Most of the reported infections caused by Photobacterium damselae subsp. damselae in humans have their primary origin in wounds exposed to salt or brackish water, inflicted during fish and tools handling. Unusual cases of infection after ingestion of raw seafood and through the urinary tract by exposure to sea water were also reported. The majority of the cases occurred in coastal areas of the United States of America, Australia, and Japan.

8Photobacterium damselae subsp. damselae can cause an extreme variant of a highly severe necrotizing fasciitis, and antibiotic administration proved unable to control the progression of fatal infections in some cases. It is interesting to note that some authors recommend to surgically debride and amputate without hesitation at a very early point of the infection by Photobacterium damselae subsp. damselae, to save the lives of patients. Some patients infected by Photobacterium damselae subsp. damselae developed multiple organ failure within a few hours from the onset of initial symptoms, despite intensive chemotherapy and surgical treatments. As an example, in a fatal case reported in 1984 in which a patient injured his hand while handling a catfish, bulle formation occurred on the hand and a marked edema extended through the forearm in less than 24 h, and although the affected area was extensively debrided the patient died after a series of complications. The bacterium was recovered in high numbers from the tissue sample but only in very small numbers from the bulle fluid. In another fatal case reported in a patient injured while handling fish, Photobacterium damselae subsp. damselae was isolated in pure culture from wound specimens but failed to be isolated from blood samples. These observations prompted these authors to suggest that a virulence factor or systemic toxin released by this bacterium contributed to the tissue damage and to the fatal outcome, rather than the septicemia itself. However, in other clinical cases this pathogen was recovered from blood.

9Necrotizing fasciitis due to Photobacterium damselae subsp. damselae demonstrates more serious complications and a higher mortality rate than that caused by Vibrio vulnificus. While V. vulnificus usually affects persons with underlying diseases (as chronic liver disease and diabetes mellitus), necrotizing fasciitis by Photobacterium damselae subsp. damselae sometimes occurs in healthy hosts.

10Early studies reported that Photobacterium damselae subsp. damselae can utilize heme, hemoglobin and ferric ammonium citrate as sole iron sources in vitro. The complete sequence of 10 genes encoding a system for the utilization of heme as iron source was described in a human isolate of Photobacterium damselae subsp. damselae, and cloning of the complete system into E. coli conferred to this species the ability to use hemin and hemoglobin as iron sources. The presence of the heme receptor gene hutA was demonstrated in subsp. damselae isolates from fish and humans, and the identity at the DNA sequence level between the heme uptake clusters of subsp. damselae and subsp. piscicida strains was 97%. Although no functional studies were conducted with the heme uptake genes of subsp. damselae, it was recently demonstrated that this cluster is essential for heme utilization in subsp. piscicida, and two genes of a hemin ABC-transporter proved to be expressed during the infective process in a fish model. Actually, an increase in the susceptibility of both fish and mice to infection by virulent Photobacterium damselae subsp. damselae strains in virulence assays conducted with iron-overloaded animals had been demonstrated in former studies. It is also known that this bacterium produces a hydroxamate-type siderophore, and the synthesis of several high-molecular weight outer membrane proteins induced under iron limitation conditions was reported. Although the precise chemical structure of the siderophore(s) is so far unknown, recent unpublished work from our laboratory demonstrated that vibrioferrin is being produced by some strains.

11Pioneering studies reported the existence of a correlation between the ability of Photobacterium damselae subsp. damselae isolates to cause disease in mice and the production of large amounts of a heat-labile cytolytic toxin in vitro. Later, the same authors purified a toxin that exhibited strong hemolytic activity against erythrocytes of a variety of animal species. In subsequent studies, this toxin named damselysin (Dly) was defined as a phospholipase-D active against sphingomyelin, with hemolytic activity, and its gene (dly) was cloned and sequenced. Dly was thus considered to be the main virulence factor of Photobacterium damselae subsp. damselae for mice. Further studies reported the existence of hemolytic strains of Photobacterium damselae subsp. damselae that tested negative for dly gene, which suggested that Dly was not the only hemolysin in the subspecies. It was also demonstrated by several authors that presence of dly is not a prerequisite for the hemolytic activity and for the pathogenicity for mice or fish, since dly negative strains bear virulence potential for animals and also toxicity for homeotherm and poikilotherm cell lines.

12The genomic context of dly gene remained uncharacterized for decades, and it was initially proposed that it could be carried on a mobile or unstable DNA element. Recently, the authors’ laboratory identified and characterized a 150 kb plasmid, which was dubbed pPHDD1, that contains the genes for Dly as well as for a HlyA toxin of the pore-forming toxin family. Only a fraction of Photobacterium damselae subsp. damselae strains harbor pPHDD1, and these strains exhibit a much wider hemolytic halo on sheep blood agar plates than the plasmidless strains (Figure 1). Interestingly, pPHDD1 occurs in both fish and human isolates and it is not restricted to a unique animal host species. In addition to being necessary to cause strong hemolytic haloes on blood agar plates, the two pPHDD1-encoded hemolysins play a crucial role in virulence for fish and mice in strains that naturally harbor the plasmid. Hence, mutation of both dly and hlyA genes in a pPHDD1-harboring strain renders the strain non-virulent for fish, and only slightly virulent for mice (Table 1), and the hemolytic phenotype on sheep blood agar of a dly and hlyA double mutant resembles that of naturally plasmidless strains (Figure 1;, ).

13The hemolytic activity exhibited by plasmidless strains was recently demonstrated to be caused by a chromosome-encoded hlyA gene, which was dubbed hlyAch in order to differentiate it from the plasmid hlyA gene (hereafter hlyApl; ). It was found that all the hemolytic Photobacterium damselae subsp. damselae strains harbor hlyAch gene, which is the only hemolytic determinant in plasmidless strains. Thus, pPHDD1-harboring isolates produce three different hemolysins. In hemolytic assays carried out with bacterial ECPs and with sheep erythrocytes, it was demonstrated that Dly acts in a synergistic manner with HlyApl and HlyAch, whereas the effect between HlyApl and HlyAch showed to be additive but not synergistic.

14Although each of the three hemolysins individually proved to be sufficient to cause death in mice, each one contributes to virulence in a different degree. The contribution of HlyAch to virulence for mice is the lowest among the three toxins. Altogether, albeit the highest values of mortality for mice are achieved only when the three hemolysins are being produced, Dly and HlyApl demonstrated to be main contributors in the virulence of Photobacterium damselae subsp. damselae for mice (; Table 1).

15Interestingly, the contribution of each hemolysin to virulence was found to vary depending on whether the host animal tested was mouse or turbot. When virulence experiments were conducted with turbot, it was found that among all the hemolysin gene mutants only the Dly-producing strains caused death in fish. This finding demonstrated that any of the two HlyA alone does not cause death in turbot, but rather one of the two HlyA needs the presence of either Dly or the other HlyA to cause death in fish. The production of Dly in combination with any of the two HlyA caused an increase in the number of dead fish with respect to the production of Dly alone, and this increase was found to be particularly evident when Dly was combined with HlyAch. This clearly suggests that, unlike what is observed in mice, the contribution of hemolysins to virulence for fish is not so much based on the individual effects of each hemolysin but rather on the combined (synergistic) effects between Dly and HlyA (; Table 1). These findings also state the importance of pPHDD1 plasmid in virulence for fish, since Dly is necessary for the synergistic effect.

16Early studies detected several enzymatic activities in the ECPs of Photobacterium damselae subsp. damselae, which included phospholipase and hemolysin activities. More recent data confirmed that the ECPs of Photobacterium damselae subsp. damselae are strongly lethal for fish, and enzymatic activities such as amylase, lipase, phospholipase, alkaline phosphatase, esterase–lipase, acid phosphatase, and β -glucosaminidase were evidenced. Moreover, treatment at 100°C for 10 min of the ECPs abolished the ability to cause death in fish, suggesting that toxicity was not due to the thermorresistant lipopolysaccharide content. Photobacterium damselae subsp. damselae ECPs also displayed cytotoxic activity for different fish and mammalian cell lines. Different studies found a correlation between virulence of the strain and toxicity of the ECPs, with toxicity being limited to ECPs from strains that were also virulent for fish. Of maximum interest is the observation that strains lacking pPHDD1 plasmid and thus being negative for dly and hlyApl genes, are virulent for fish and their ECPs are cytotoxic for cell lines. In addition, a comprehensive study reported that none of the enzymatic activities detected in the Photobacterium damselae subsp. damselae ECPs could be related with the degree of toxicity either in vivo or in vitro. Most Photobacterium damselae subsp. damselae strains test negative for protease activities as caseinase and gelatinase. This suggests that other, yet uncharacterized molecules produced by Photobacterium damselae subsp. damselae cells play a role in toxicity for animals and for cell lines. In this regard, previous studies detected the existence of an acetylcholinesterase activity (ictiotoxin) with neurotoxic activity in several species of Vibrionaceae, including Photobacterium damselae subsp. damselae, although the genetic basis for this neurotoxic activity remains unknown.

17An interesting observation that remains to be explained at the genetic level is the finding that plasmidless Photobacterium damselae subsp. damselae strains are virulent for fish and toxic for homeotherm and poikilotherm cell lines. Since plasmidless strains lack dly and hlyApl genes, and since dly hlyApl double mutants are significantly reduced in its virulence for both mice and fish, it is evident that plasmidless strains encode virulence factors that either are not encoded by pPHDD1-harboring strains or their expression is repressed in presence of pPHDD1-encoded genes.

18The recent completion of the genome sequence of the type strain (ATCC 33539) of this subspecies (deposited in GenBank database in several separate contigs, under accession number ADBS00000000), allows an in silico analysis to search for candidate genes encoding potential toxins and other virulence factors. The type strain harbors pPHDD1 plasmid, and preliminary analyses also indicated the presence of genes encoding a type III hemolysin (open reading frame number: VDA003208), and a putative murine toxin (VDA000322) among others. The existence of yet uncharacterized plasmids is also evidenced in the complete genome of ATCC 33539. Studies to functionally characterize novel plasmid content and candidate virulence genes of Photobacterium damselae subsp. damselae strains are currently under way. It is expected that a deep analysis of the complete genome sequence of Photobacterium damselae subsp. damselae strains with different isolation origins and virulence properties will provide a clearer picture of the virulence factors employed by this bacterium to cause disease in such a varied range of hosts.

19The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.