Aquaculture, breeding, and rearing of aquatic animals (mollusks, crustaceans, and
finfish) in marine, brackish, and freshwater bodies is playing an increasingly important
role with respect to food security for the growing human population and is predicted
to dominate the seafood supply within a few decades [1]. However, the further sustainable
expansion of the sector is currently hampered by a number of factors, amongst which
diseases are playing a prominent role, especially in the early life stages of the
animals (i.e., larviculture [2]). A major group of causative agents are bacterial
pathogens, such as Vibrio spp. [3]. These bacteria cause huge losses in the aquaculture
industry worldwide, with acute hepatopancreatic necrosis disease (AHPND) as a notable
recent example [4]. This disease, caused by strains of Vibrio parahaemolyticus that
acquired a plasmid encoding two toxin genes, was first reported in southern China
in 2009,subsequently spread over Southeast Asia, and reached Mexico in 2013. The AHPND
disease typically affects shrimp postlarvae, within 20–30 days after stocking, and
frequently causes up to 100% mortality. Global losses in the shrimp farming industry
because of this disease have been estimated to be over US$1,000 million per year [5].
Pathogenic vibrios are opportunistic pathogens (as opposed to obligate pathogens)
since they are capable of surviving and multiplying in the absence of their host [6].
In a recent paper, De Schryver and Vadstein argued that the ecological r/K theory
could serve as a foundation for the development of microbial management strategies
to prevent diseases caused by opportunistic pathogens in aquaculture [7]. According
to the ecological theory of r/K selection, an unstable environment containing high
nutrient levels per individual selects for organisms with the capacity to exploit
nutrients and increase population size, termed r-strategists. On the other hand, a
stable environment where the resources per individual are scarce will select for slow-growing
organisms, termed K- strategists. Most bacterial diseases in aquaculture (and especially
larviculture) are caused by opportunistic pathogens that are ubiquitous in the marine
environment and that are capable of quickly increasing their population size in the
aquaculture environment: i.e., r-strategists [2,7]. The triggers that induce mortality
events are not yet completely understood (and are probably different for different
pathogens), although increases in dissolved nutrients, temperature, and periods of
hypoxia have been shown to be involved [8–10].
Management strategies aiming at disease prevention in aquaculture systems should be
pursued at different levels, the first of which includes the implementation of hygienic
barriers (e.g., the disinfection of incoming water in order to avoid those pathogens
that are part of the normal marine microbiota from entering the system). These hygienic
barriers are, however, not flawless and do not result in a complete eradication of
all incoming bacteria, and consequently, additional measures should be taken to restrain
pathogens within the system [11]. Moreover, disinfection practices can favor r-strategists
because they decrease the competition between bacteria (by decreasing the bacterial
density), and this also holds for feeding regimes that result in large fluctuations
in nutrient levels in the rearing water [12]. Given the fact that many of the major
bacterial aquaculture pathogens are r-strategists, theoretically, by imposing slow
growth conditions in order to favor K-strategists (i.e., organisms capable of thriving
in an environment with low levels of nutrients per individual), the pathogen pressure
in the aquaculture system can be decreased, and this should result in a lower incidence
of diseases [7]. Slow growth conditions can be imposed by controlled microbial colonization
of the (disinfected) inflow water with K-strategists (matured water [13]) and by avoiding
large fluctuations in nutrient levels in the water—e.g., by imposing a feeding regime
consisting of a continuous administration of low feed doses instead of a regime consisting
of few feeding events in which relatively large amounts of feed are introduced in
the system. The reasoning behind this is that dissolved nutrient levels will be kept
low because of consumption by the microbiota in the matured water, thereby minimising
opportunities for opportunistic r-strategists to invade the system. This approach
has been experimentally validated (e.g., in Atlantic cod larval rearing, in which
a matured water approach resulted in an increased survival [14]). However, the success
seems to be variable, and survival rates in systems with matured water are often still
relatively low (although higher than in other systems). Hence, although microbial
management based on slow growth to favor K-strategists is intellectually appealing
and to some extent supported by experimental data, there still seems to be room for
improvement. In the following paragraphs, I will argue that microbial management strategies
need to take into account the fact that nutrients are not homogeneously distributed
in the water column. Indeed, recent findings with respect to ecological niche differentiation
of marine bacteria indicate that there still is an open window for invasion by opportunistic
pathogens in matured water that has been colonised by K-strategists because these
pathogens have evolved mechanisms to find and exploit hot spots with high nutrient
levels [15,16].
Despite its superficial homogeneous appearance, the marine water column can have a
diverse physical, chemical, and biological microenvironment, and nutrients are not
homogeneously distributed at scales relevant to microorganisms but rather occur as
hot spots, as they are, e.g., often associated with or released from particles such
as microalgae, fecal pellets, and marine snow [16]. In order to cope with the conditions
that prevail in the marine environment (i.e., low nutrient levels in the bulk and
high levels in hot spots), marine bacteria have evolved two divergent strategies:
they are either (1) minute and nonmotile, with a streamlined genome, or (2) relatively
large and motile, with a high metabolic flexibility [16]. The small cell size of the
first group allows them to maximise uptake per unit of biomass and to obtain nutrients
at the low bulk concentrations of the oceans [17]. Their streamlined genomes imply
poor metabolic plasticity and an inability to exploit high-resource conditions that
occur in hot spots [18]. The second, motile group is adapted to exploit these relatively
rare, resource-rich conditions. Chemotactic motility enables them to access novel,
nutrient-rich hot spots [16]. Their metabolic flexibility allows them to adapt rapidly
to newly encountered microenvironments [19]. Interestingly, the division between both
strategies described above is broadly aligned with the dichotomy between K-strategists
and r-strategists, and the abundance of r-strategists will be a reflection of the
patchiness of the water column [16]. The Vibrionaceae family contains the major bacterial
pathogens in marine aquaculture, and they belong to the second group: they are often
(highly) motile, show a high metabolic flexibility, and are capable of quickly increasing
their population size in the environment if conditions are favourable [16].
Similar to oceans, the water column of aquaculture systems is not homogeneous, and
nutrients are usually also present in hot spots (fecal pellets, uneaten feed particles),
which are highly abundant in the rearing water of an aquaculture system (at least
when compared to the presence of hot spots in the ocean). Hence, there are plenty
of opportunities for r-strategists (including opportunistic pathogens) in aquaculture
systems. If nutrients were uniformly distributed, then matured water (containing a
high level of K-strategists) would adequately limit the risk of invasion by opportunistic
pathogens (Fig 1A). However, the fact that nutrients are not uniformly distributed
includes a significant risk for invasion by opportunistic pathogens, even in matured
water, because they have a competitive advantage over K-strategists with respect to
obtaining nutrients from hot spots (and thereby increasing their population density)
(Fig 1B). Along this line, Lemire et al. recently reported that oyster pathogenic
vibrios belong to distinct ecological populations that show preference for zooplankton
and large particles [20].
10.1371/journal.ppat.1005843.g001
Fig 1
Schematic representation of the impact of nutrient distribution on the opportunities
for an opportunistic pathogen to invade the preexisting microbial community in an
aquaculture system (matured water).
The darker the colour, the higher the nutrient concentration. (A) In the hypothetical
scenario in which nutrients are homogeneously distributed in the water column, K-strategists
are capable of using the nutrients, and there is no opportunity for the opportunistic
pathogen to reach a high population density. (B) In reality, nutrients are not homogeneously
distributed but rather present as hot spots with locally high nutrient concentrations.
The nonmotile K-strategists remain randomly distributed and are not able to adequately
exploit the nutrient source. Chemotactic motility enables the opportunistic pathogen
to localise and exploit the hot spot, and as a consequence, its population density
increases considerably. (C) If the original microbial community of the matured water
also contains r-strategists, then these r-strategists can compete with opportunistic
pathogens for the resources present in the hot spots. As a consequence, the opportunistic
pathogen is not able to reach a density as high as in panel B. (D) If feed particles
are used that disintegrate rapidly, then the time window during which the hot spot
exists is limited and nutrients quickly diffuse into the bulk, thereby avoiding (locally)
high levels of dissolved nutrients. As a consequence, K-strategists will be capable
of utilising a significant fraction of the nutrients, and the opportunistic pathogen
is not able to reach a density as high as in panel B.
Although ecological niche differentiation in marine bacteria includes a risk for invasion
of aquaculture systems by opportunistic pathogens, further research is needed in order
to fully appreciate the magnitude of this risk and to develop novel microbial management
strategies to limit the risk of invasion by opportunistic pathogens in an aquaculture
system. Such a strategy could consist of including nonpathogenic r-strategists in
the matured water (Fig 1C). Indeed, in this way, the rearing water would also contain
bacteria that are able to occupy the ecological niche that is prone to invasion by
opportunistic pathogens. It is clear that in this case, r-strategists should be carefully
selected—e.g., taking into account the fact that within the Vibrionaceae, ecological
population boundaries can be at a low phylogenetic level (e.g., several ecologically
distinct populations could be distinguished within the species V. splendidus [21]).
On the other hand, recent work indicates that a large fraction of strains within a
certain ecological population (although not all of them) can be pathogenic [21], and
consequently, it might prove to be challenging (though not impossible) to identify
appropriate r-strategists that could be used as inoculum for the precolonization of
the intake water. An additional strategy is to use feed pellets that disintegrate
relatively quickly, as this will decrease the window during which r-strategists have
an advantage over the K-strategists present in the matured water (Fig 1D). Indeed,
if the pellets disintegrate quickly, then the time frame during which local dissolved
nutrient levels are high will be limited, as the nutrients will diffuse more quickly
into the bulk water, where they can be consumed by K-strategists. It needs to be stressed,
however, that these pellets should be used in combination with a slow feeding regime
(see above) in order to avoid temporarily high levels of dissolved nutrients in the
water.