Diagnosis and Control of Diseases of Fish and Shellfish

List of Contributors

Brian Austin

University of Stirling

Scotland

UK

Mags Crumlish

Institute of Aquaculture

University of Stirling

Scotland

UK

Andrea Gustinelli

Department of Veterinary Medical Sciences

Alma Mater Studiorum

University of Bologna

Bologna

Italy

Seyed Hossein Hoseinifar

Department of Fisheries

Gorgan University of Agricultural Sciences and Natural Resources

Gorgan

Iran

Ahran Kim

Department of Aquatic Life Medicine

Pukyong National University

Busan

Republic of Korea

Do-Hyung Kim

Department of Aquatic Life Medicine

Pukyong National University

Busan

Republic of Korea

Matt Longshaw

Benchmark Animal Health Ltd

Edinburgh Technopole

Scotland

UK

Thanh Luan Nguyen

Department of Aquatic Life Medicine

Pukyong National University

Busan

Republic of Korea

Simon MacKenzie

Institute of Aquaculture

University of Stirling

Scotland

UK

Aweeda Newaj-Fyzul

University of the West Indies

Trinidad

and

Agriquatics

Chaguanas

Trinidad

Giuseppe Paladini

Institute of Aquaculture

University of Stirling

Scotland

UK

Miriam Reverter

CRIOBE, Paris Sciences et Lettres (PSL)

University of Perpignan Via Domitia

Perpignan

France

and

Laboratoire d'Excellence CORAIL

Moorea

French Polynesia

Felipe Reyes-López

Department of Cell Biology

Physiology and Immunology

Universitat Autonoma de Barcelona

Bellaterra

Spain

Pierre Sasal

CRIOBE, Paris Sciences et Lettres (PSL)

University of Perpignan Via Domitia

Perpignan

France

and

Laboratoire d'Excellence CORAIL

Moorea

French Polynesia

Denis Saulnier

Laboratoire d'Excellence CORAIL

Moorea

French Polynesia

and

Ifremer

Taravao

Tahiti

French Polynesia

S.M. Sharifuzzaman

Institute of Marine Sciences and Fisheries

University of Chittagong

Bangladesh

Andrew P. Shinn

Benchmark Animal Health Ltd

Edinburgh Technopole

Scotland

UK

and

Fish Vet Group Asia Ltd

Chonburi

Thailand

Yun-Zhang Sun

Key Laboratory of Healthy Mariculture for the East China Sea

Ministry of Agriculture

Fisheries College

Jimei University

Xiamen

China

Nathalie Tapissier-Bontemps

CRIOBE, Paris Sciences et Lettres (PSL)

University of Perpignan Via Domitia

Perpignan

France

and

Laboratoire d'Excellence CORAIL

Moorea

French Polynesia

Eva Vallejos-Vidal

Institut de Biotecnologia i Biomedicina

Universitat Autonoma de Barcelona

Bellaterra

Spain

Manfred Weidmann

Virology Unit

Institute of Aquaculture

University of Stirling

Scotland

UK

Zhigzhang Zhou

Key Laboratory for Feed Biotechnology of the Ministry of Agriculture

Feed Research Institute

Chinese Academy of Agricultural Sciences

Beijing

China

Preface

There has been continual expansion in aquaculture since the end of the Second World War, and currently total production is approximately equal to that of wild-caught fisheries. Yet, this expansion is marred by continued problems of disease. New pathogens emerge, and others become associated with new conditions. Some of these pathogens become well established and develop into major killers of aquatic species. Examples include infectious salmon anaemia virus, francisellosis and amoebic gill disease. Research has seen significant developments in diagnostics and disease control. The former has progressed from the descriptive, through serological to molecular methods. This progression has led to greater sensitivity, specificity and reliability. However, it is not always clear what a positive result means in the absence of clinical disease manifestations. Developments in disease control have encompassed therapy (disinfectants, antibiotics) and prophylaxis (vaccines, probiotics, prebiotics, immunostimulants). There is a trend away from the use of chemicals because of issues with the development and spread of resistance, tissue residues and environmental concerns. Interest in disease prevention has soared, from the development of vaccines to interest in probiotics, and now to the use of medicinal plant products.

The primary aim of this book is to focus on developments in the diagnosis and control of diseases of fish and shellfish, notably those affecting aquaculture. The book is primarily targeted at research workers, including postgraduate students, diagnosticians and individuals concerned with the management of diseases of fish and shellfish. It is anticipated that the readership will include veterinarians, fish pathologists, microbiologists, public health scientists and microbial ecologists.

Brian Austin and Aweeda Newaj-Fyzul

2017

Chapter 1

Introduction

Brian Austin¹ and Aweeda Newaj-Fyzul²

¹Institute of Aquaculture, University of Stirling, Stirling, Scotland, UK

²University of the West Indies, Trinidad and Agriquatics, Chaguanas, Trinidad

There is confusion over the meaning of the term ‘disease’. A definition from an article in the British Medical Journal is as follows:

… a disease is the sum of the abnormal phenomena displayed by a group of living organisms in association with a specified common characteristic or set of characteristics by which they differ from the norm of their species in such a way as to place them at a biological disadvantage. (Campbell et al., 1979)

According to these authors, a disease is something that occurs to a group of organisms rather than to an individual. Also, the definition is far reaching and reflects the complex relationship between the disease-causing situation (not necessarily a micro-organism) and the host. However, there is more to disease than the interaction of a pathogen (dictionary definition is of a disease-causing organism) and the host. According to Kinne (1980), who was writing about diseases of marine animals, diseases could be caused by genetic disorders, nutritional imbalance, pathogens, physical injury and pollution. Thus, disease could be attributed to biological (biotic) as well as non-biological (abiotic) causes. Kinne (1980) described diseases in terms of epidemiology (epizootiology for animal diseases), as follows.

Sporadic diseases, which occur sporadically in comparatively small numbers of individuals in a population.

Epidemics/Epizootics, which are large-scale outbreaks of communicable disease occurring temporarily in limited areas.

Pandemics/Panzootics, which are large-scale outbreaks of communicable disease occurring over large areas.

Endemics/Enzootics, which are diseases persisting or reoccurring as low-level outbreaks in defined areas.

The interest in diseases of aquatic organisms is primarily directed towards aquaculture which, to paraphrase definitions, is the rearing of aquatic species in controlled conditions. Here, disease may be of sudden onset with rapid progression to high mortalities, with an equally quick decline (acute disease). Conversely, there may be cases where the disease develops more slowly, with less severity but longer persistence (chronic disease).

It is apparent that as society moves through the twenty-first century, aquatic animals continue to suffer the vagaries of disease, especially as new diseases continue to occur, e.g. acute hepatopancreatic necrosis disease syndrome (AHPND), which is attributed to infection of the shrimp with Vibrio parahaemolyticus. However, the study of aquatic pathobiology is largely an aerobic affair as most laboratories do not consider the possible role of anaerobes or microaerophiles. It would be interesting to determine if this reflects a lack of expertise, interest or suitable methods rather than the absence of occurrence of anaerobic or microaerophilic pathogens/parasites. The majority of literature points to a single species of pathogen as the main cause of disease situations but there are reports of sequential viral and bacterial as well as parasitic and opportunistic bacterial combinations. Certainly, it is appreciated that there have been important developments in disease diagnosis, with progression from traditional phenotyping to the use of newly developed molecular techniques. In many modern laboratories, identification is now routinely accomplished by sequencing of the 16S rRNA gene, a move that has led to greater confidence in the outputs although this will reflect the accuracy of the data in the databases.

However, whereas the use of new technologies is to be encouraged, an ongoing dilemma remains about the authenticity of isolates studied. Also, many studies are based on the examination of single isolates, the relevance of which to fish pathology or science in general is not always clear. Certainly, too many conclusions result from the examination of too few isolates. Nevertheless, the study of pathogenicity mechanisms, diagnostics and disease control by means of vaccines has benefited from molecular approaches. Yet it is appreciated that many laboratories still rely on conventional methods for achieving disease diagnoses, and it is unlikely that this situation will change in the foreseeable future. However, it is pertinent to enquire what diagnosis is supposed to achieve. If the underlying aim is to underpin efforts at disease control, then it is unclear whether detailed and time-consuming work resulting in identifying a pathogen to species level would necessarily help disease control strategies.

The use of pathological specimens taken from advanced cases of disease would be unlikely to reveal species succession, which could occur throughout a disease cycle. Also, we focus on pure cultures, and are generally uncomfortable with the notion that two or more organisms could be associated with a single disease condition. Diagnostic microbiology aims to isolate the dominant organism, as a single pure culture, from pathological material. It is speculative how often the wrong organism may be chosen, as the true pathogen becomes overwhelmed by contaminants. Yet laboratory cultures are used extensively for associated studies of pathogenicity mechanism and disease control. Unfortunately, all too often, cultures lose some of their characteristics in the laboratory which may reflect loss of DNA – and therefore ensuing data need not reflect the true role of the culture with its host.

Histological examination of diseased tissue may be invaluable in recognizing cases where organisms, presumed to the pathogen, may be observed but culturing not achieved. The uncultured Candidatus have become associated with some diseases. It is unclear if such organisms are incapable of growing in vitro or if suitable media have not been developed. It is unknown how many more of these unculturable organisms remain to be recognized. Could such uncultured organisms be dormant, damaged or senescent, a concept which has been put forward for some water-borne organisms by Stevenson (1978)? Then there are situations where there is pathology, for example red mark syndrome of rainbow trout, for which the cause is uncertain but may reflect infection with a rickettsia (Metselaar et al., 2010).

It is interesting to note that Al Gore, when Vice President of the USA, suggested that diseases (of humans) would be controlled within our lifetimes. Since then, human society has suffered from the emergence of new diseases (e.g. Middle East respiratory syndrome (MERS) coronavirus) and the recurrence of others (e.g. Ebola, H1N1 and Zika). The use of medically important antibiotics in any non-medical situation, including aquaculture, is fraught with problems, of which the development and spread of resistance, and issues with tissue level, top the list of concerns. The result is that there is a deliberate move away from the use of antibiotics in many countries. Certainly, control of diseases of aquatic organisms has undergone dramatic improvements from the initial emphasis on control (therapy) to prevention (prophylaxis). Unfortunately, the dramatic progress in human vaccinology has not been reflected in the number of commercial products available to aquaculture. However, there is no shortage of ingenuity in the development of disease control strategies, as illustrated by the growing interest in non-specific immunostimulants, prebiotics, probiotics and plant products.

This text will focus on the diagnosis and control of infectious diseases of farmed aquatic animals.

Conclusion

The list of parasitic, bacterial and viral pathogens continues to grow, although the significance of some organisms to pathology is difficult to ascertain – are they truly parasite/pathogens, secondary invaders or contaminants?

There has been considerable improvement in the taxonomy and hence diagnosis (accuracy and sensitivity) of many pathogens, particularly involving the sequencing of the 16S rRNA gene.

Disease control has progressed from therapeutic to prophylactic, and now involves a wide range of approaches including vaccines, non-specific immunostimulants, prebiotics, probiotics and plant products.

References

Campbell, E.J.M., Scadding, J.G. and Roberts, M.S. (1979) The concept of disease. British Medical Journal, 2, 757–762.

Kinne, O. (1980) Diseases of Marine Animals, vol 1. General Aspects, Protozoa to Gastropoda, John Wiley & Sons, Chichester.

Metselaar, M., Thompson, K.D., Gratacap, R.M.L., et al. (2010) Association of red-mark syndrome with a Rickettsia-like organism and its connection with strawberry disease in the USA. Journal of Fish Diseases, 33, 849–858.

Stevenson, L.H. (1978) A case for bacterial dormancy in aquatic systems. Microbial Ecology, 4, 127–133.

Chapter 2

Bacterial Diagnosis and Control in Fish and Shellfish

Mags Crumlish

Institute of Aquaculture, University of Stirling, Stirling, Scotland, UK

Introduction

Aquaculture is described as an ‘organised production of the crop in the aquatic medium’ (FAO, 1987) and while this might be considered a very generic description, it is actually quite accurate given the diversity in production and range of species farmed. These systems are often categorized as either intensive, semi-intensive or extensive which is more a reflection of the production levels in these types of systems as the boundaries between each category are not well defined. Aquaculture is practised in all parts of the world and farms may be located in coastal areas, open marine water and inland in earthen ponds or river-based cages. Given the ubiquitous presence of bacteria within the environment, disease outbreaks can occur in each of these production types and locations.

Global aquaculture production has continued to expand at approximately 9% per year since the 1970s, with Asia dominating the production levels, particularly in finfish (FAO, 2014). It is in Asia that we see the largest and most rapid expansion. The growth of Asian aquaculture has outstripped that of European and North American production, which remains high but with limited capacity for significant growth compared with Asian aquaculture (Bostock et al., 2010). This is primarily due to the increasing desire for intensification but also the availability of the more diverse species range suitable for farming in Asia. At present, it is estimated that more than 600 aquatic species are raised in freshwater, brackish and marine farms of varied intensity (FAO, 2014). These 600 species include both vertebrate and invertebrate animal species as well as plants, but for the purposes of this chapter, we will focus our attention on the most intensive farmed species that are traded globally for human food and include examples from finfish and crustaceans only. From this point onwards, the use of the term ‘fish’ includes finfish and shrimp unless otherwise stated.

Bacterial Infections in Aquaculture

Bacteria are described as single-celled organisms that have a rather simple cellular structure, lacking membrane-bound organelles. They are found ubiquitously in all habitats, including fresh and sea water, and display a range of cellular morphologies. Bacterial classification relies on identification of phenotypic and genotypic characteristics and the relatively simple Gram stain reaction remains the most reliable method allowing species to be separated into either Gram-positive or Gram-negative groups. It is the chemical and physical properties of the bacterial cell wall that allow the retention of the coloured dye used during the Gram stain reaction. In aquaculture, disease outbreaks occur from both Gram-negative and Gram-positive bacterial species, which may be rod-like or spherical cocci in shape. It is not the purpose of this chapter to discuss in detail the varied bacterial species; the reader is referred to Austin and Austin (2016) for more in-depth detail on specific aquatic bacterial pathogens. A list of the commonly reported bacterial diseases that affect intensive monoculture systems is provided in Table 2.1.

Table 2.1 Bacterial pathogens commonly reported in intensive production systems

The bacterial pathogens that have been identified and characterized the most are those that cause greater economic impact as determined through high mortalities or morbidity at the farm. At present, it is fair to say that we see more infections from Gram-negative than Gram-positive species (see Table 2.1). However, further intensification and introduction of novel host species combined with increasing consumer demand for non-local or exotic food types may change this in the future. Furthermore, bacterial identification and taxonomy is a rapidly developing area (Austin and Austin, 2016) as we move away from phenotypic-only tests and rely more on molecular tools for pathogen identification. This will not only result in taxonomic changes but with appropriate development, such methods may provide additional diagnostic tools leading to the production of novel control strategies applicable within aquaculture.

Bacterial Disease Diagnostics and Control of Infections

The principles behind aquatic bacterial diagnostics are similar to those practised in human clinical and terrestrial veterinary medicine. While the methods and approaches are similar, the type of samples and the diagnostic tests used will depend on the reason for the initial investigation. In aquaculture, diagnostic samples are provided to the laboratory to determine the health status of animals prior to transportation of live shipments or used to confirm that animals are specific pathogen free (SPF). However, like other farming sectors, the most common use of diagnostics in aquaculture is investigation of an unexpected mortality or morbidity within the farmed stocks from a suspected disease. Not all causes of mortality are infectious and so disease outbreaks can only be confirmed using a diagnostic approach. This means that to perform the diagnosis, we need to have a combination of information which includes the farm history and outbreak or event history, followed by a visual examination of the animals with and without clinical signs prior to taking samples for the laboratory tests.

Disease outbreaks are multifactorial, where the clinical outcome is dependent on the interaction between the host and the pathogen. To be more accurate, it is the interaction of the host immune response with the virulence factors produced by the pathogen that provides the range of clinical signs observed. Disease outbreaks in aquatic farms are often described as either acute or chronic, which is a reflection of the onset of the disease condition rather than an accurate description of the infection itself.

Reliance on observations of gross clinical signs of disease in aquaculture is limited, as the clinical presentation can vary tremendously and not all clinical signs have a microbial aetiology. Infections due to the Gram-negative bacterium Edwardsiella ictaluri in Asian catfish species Pangasianodon hypophthalmus provide few if any external clinical signs (Ferguson et al., 2001). It is only upon internal examination of these fish that the clinical signs may become more apparent. However, the clinical signs associated with infections due to Aeromonas hydrophila in many freshwater farmed fish species can be grossly similar to pathology arising from poor handling or grading stress. Identification of the cause of the disease problem is not simple as many factors, including both biotic and abiotic, can contribute to the disease process. It is fair to say that we are still scratching the surface of knowledge about the disease dynamics involved in an infectious disease outbreak within the aquatic farming environment.

In aquaculture, like other disciplines, the role of the bacteriology laboratory is to screen samples provided and identify the aetiological agent causing the disease under investigation. Most samples submitted to the laboratory are in the form of inoculated media plates which are incubated at an appropriate temperature and examined for colony growth and purity of culture. Given that recovery of a single bacterial species is rare from farmed aquatic animals, especially those that are sick, care must be taken during sampling to control unwanted contamination. The purpose of the initial colony morphology screening on the agar plates is to support the diagnosis of an infectious aetiology and not to identify every single organism that is recovered on the media. There is also a need to ensure that all sampling and subsequent investigations are performed aseptically to avoid unwanted contamination.

Species identification has traditionally relied on phenotypic tests to provide the identification of the suspected pathogen (Frerichs and Millar, 1993). It is the prerogative of each individual laboratory to decide on the number of samples to process and the range of assays to include during pathogen identification. This will depend on the capacity and capabilities of the laboratory. Primary identification tests often include Gram stain, a motility test, identification of either oxidase and/or catalase enzyme reaction and determination of oxidation or fermentation of a carbohydrate substrate such as glucose (Frerichs and Millar, 1993). This is usually sufficient to identify the bacterial species to genus if not species level and further testing can include biochemical profiling using one of the many commercially available kits.

The value of such traditional primary bacterial identification assays has been reduced over time as we include more molecular tools and yet, in many cases, they can provide immediate results and are cheap to perform. The cost-effectiveness of any test is important within any diagnostic service. Several biochemical kits are commercially available which are not specifically designed for use with aquatic pathogens but several authors have found them useful after minor adaptations. There are several types of kits available but the most frequently used include the API 20E, API ZYM, API 20NE, API 50 CH, Vitek® system (bioMérieux) and Biolog MicroPlates (Biolog, Inc.) (Popovic et al., 2007).

All identification tests take place in the laboratory under controlled conditions following standard procedures as provided by the diagnostic staff and facility. Such protocols would be tested for sensitivity and specificity prior to implementation and should be reviewed regularly to ensure the most suitable methods are being applied. As in other animal production sectors, we have seen a move away from the more traditional bacterial identification methods towards the application of molecular-based probes, using nucleotide sequence analysis (Austin and Austin, 2016). The development of new technologies within aquatic diagnostics has relied on adaptations of techniques applied within human clinical and veterinary medicine (Adams and Thompson, 2011) and has led to the development of ‘rapid diagnostics’.

Modern Approaches in Bacterial Diagnostics

Conventionally, most bacterial diagnostic assays are laboratory based but there is a move towards the production of more rapid diagnostic tools that can be applied directly at the farm site. The on-farm application is attractive within aquaculture, particularly in intensive farming practices where there are thousands of animals in a single farm. Therefore, one of the strengths of the pond or cageside diagnostic kits may lie in the ability to screen large populations for the presence of specific pathogens or the more rapid identification of fastidious organisms such as Flavobacterium psychrophilum, the aetiological agent of bacterial cold-water disease and rainbow trout fry syndrome (Davis, 1946; Madsen and Dalsgaard, 1998). Before moving away from centralized laboratories, first we need to develop a robust and reliable kit that is easy to use and provides rapid results at the farm. This might be considered the Holy Grail of aquatic bacterial infections.

Traditional laboratory-based identification tests are often considered time consuming and labour intensive. This is not productive within a diagnostic service where the emphasis lies on the ability to confirm the diagnosis in a timely manner. This has led to the development and uptake of molecular probes within aquatic disease diagnosis. Most bacteriology laboratories will use or have access to polymerase chain reaction (PCR) assays to help in the identification of specific bacterial pathogens. The PCR-based technologies rely on the detection of a common target specific to the bacterial species. This is normally within the conserved genes present on the bacterial ribosomal RNA or it may be detection of a specific housekeeping gene (Frans et al., 2008). Over the last 20 years, numerous 16S rRNA assays have been developed and can be used for the identification or confirmation of a bacterial species within a sample. The obvious benefit of molecular tools is time reduction, especially as these assays are not reliant on the recovery of viable cultures.

The trend in application of nucleic acid assays within aquatic diagnosis has led to the development of multiplex PCR which allows the detection of multiple targets (pathogens) in a single sample. These are more advantageous as detection of more than one target in a single sample can significantly reduce the amount of time and consumables used when processing, thus reducing the overall assay costs. The lack of commercially available vaccines to control aquatic diseases is one of the key drivers in the development of rapid diagnostic kits.

Multiplex PCR assays have almost become routine within aquatic animal health research, with many providing simultaneous detection of three or more major pathogens within the sample. These have been developed for freshwater bacterial pathogens (del Cerro et al., 2002; Panangala et al., 2007), marine bacterial species (Gonzalez et al., 2004) and several species of Gram-positive cocci causing streptococcosis in farmed fish (Mata et al., 2004). Of course, all bacterial strains are not equally pathogenic, which led to the implementation of gene-specific DNA microarray assays that differentiated pathogenic strains. This was particularly useful in the detection of pathogenic Vibrio species within shellfish studies as the microbial community is likely to consist of a combination of pathogenic and non-pathogenic strains present within the sample (Panicker et al., 2004).

Microbial epidemiology studies have applied the varied bacterial DNA fingerprinting methods, such as multilocus sequence typing (MLST) or pulsefield gel electrophoresis (PFGE), to promote rapid identification of infectious bacterial strains. Such assays have often relied on the presence of several ‘housekeeping’ genes and compared large numbers of strains from varied clinical outbreaks (Delannoy et al., 2013). Screening large numbers of clinically distinct isolates provides the most robust data in understanding the incidence and prevalence of disease outbreaks. While such approaches have been used in aquatic pathogen studies, they remain expensive and time consuming and hence are not readily applicable for rapid diagnosis as yet. Whole genome sequencing was once considered prohibitively expensive but the costs are rapidly reducing and this is becoming a more attractive tool to support advances in aquatic bacterial disease investigation. It is in the field of proteomic research that perhaps we see the future of bacterial identification and disease diagnosis. The use of matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) allows the identification of bacteria using either intact cells or cell extracts. It is described as rapid, highly sensitive and economically cost-effective in terms of labour and running costs (Singhal et al., 2015). The application of MALDI-TOF MS within the fields of clinical medicine and food microbiology is rapidly advancing (Pavlovic et al., 2013) although uptake within aquatic bacterial studies is more limited.

Most aquaculture is performed in rural locations which is why the move towards pondside or on-farm rapid diagnostic kits is particularly attractive, as they can be applied on site. Lateral flow kits are good examples of the novel methods being developed in aquaculture. In the field of human medicine, such assays can be widely implemented, give a rapid response and support healthcare and diagnostics. The reader is directed to a review by Sajid et al. (2015) which provides excellent detail on the varied designs, formats and potential applications of lateral flow assays generally.

There are several research papers which advocate the use of rapid diagnostic kits in aquaculture and yet industry uptake has been poor. Technology transfer from the laboratory to the field has been a much more arduous process. Prior to implementation within routine diagnostics, more evidence is needed on how such methods might advantage the diagnosis where the clear benefit would be in early detection of disease outbreaks leading to measurable reduction in animal losses. There is a recognized need to ensure accuracy, specificity and sensitivity of the particular assay or kit but, perhaps more importantly, in aquaculture there is the cost-benefit analysis to consider.

Control Strategies Against Bacterial Diseases

Intensive farming systems have a health management plan which is embedded within the farming practice and will encompasses optimal nutritional, water quality and husbandry to ensure the health and welfare of the animals during production. Broadly, the health management plan will include prevention of disease outbreaks and treatment regimes during infections. Vaccination is now considered routine practice for many intensive finfish aquaculture systems where greater development of attenuated and DNA vaccines is predicted (Brudeseth et al., 2013). However, the lack of commercially available vaccines for the global finfish culture as a whole means that there is a perpetual reliance on antibiotics to treat bacterial infections. Furthermore, farmed shrimp species cannot be conventionally vaccinated as they lack the appropriate immune system (Xiong et al., 2016). This leads to greater reliance on biosecurity and alternative chemotherapeutants.

Biosecurity can be considered as a set of criteria, which may include physical, chemical or biological variables designed to protect against the entry and spread of pathogens within the farming system. As aquaculture has grown and intensified, we have seen a shift away from a comprehensive biosecure plan towards a more pathogen-specific approach. In some aquaculture sectors, this may be economically beneficial but the lack of a comprehensive biosecurity plan can leave the farming sector vulnerable to emerging aquatic diseases. A good example of this is the recent outbreak of acute hepatopancreatic necrosis disease, also called early mortality syndrome, due to specific strains of the bacterium Vibrio parahaemolyticus (Tran et al., 2013). This disease was first reported in warm-water shrimp farms in China in 2009 and quickly spread to other countries which were intensively farming the warm-water shrimp (Zorriehzahra and Banaederakhshan, 2015). The impact of this disease heavily affected the global supply of warm-water farmed shrimp species (Penaeus monodon and Litopenaeus vannamei) where at the peak of the outbreak in 2013, global annual losses of more than US$1 billion were reported (GAA, 2013). Over-reliance on a single biosecurity practice such as only stocking with specific pathogen-free shrimp may weaken the overall biosecurity practices on the farm, leaving the sector vulnerable to the emergence of new pathogens or diseases.

By following a therapeutic regime, we can use antibiotics to effectively treat a bacterial infection in aquatic farming systems. Unfortunately, the concept of therapeutic antibiotic treatments has not been widely used outside the intensive salmon sector. There are many reasons for this but the lack of rigour and frequent misuse of antibiotics in aquaculture have promoted the spread of antibiotic resistance (Defoirdt et al., 2011). Alternatives are actively being sought which has reignited the interest in bacteriophage therapy and probiotics (Newaj-Fyzul and Austin, 2015) in aquaculture.

Emerging Bacterial Diseases

Previously we have mentioned the growth and intensification of the global aquaculture sector and highlighted these strengths in enabling farmed aquatic food to become a significant contributor within global food security. However, without proper vigilance, these strengths may become opportunities for future disease outbreaks and certainly we are seeing a worrying trend in new and emerging diseases in aquaculture. Emerging diseases are not limited to the presence of novel pathogens but can also be applied more widely to mean appearance of an existing disease in a new geographical location or increased incidence of the disease. While new diseases continue to emerge, there are those that have had a more immediate impact in the global aquaculture sector. Outbreaks of francisellosis have caused significant mortalities in both warm- and cold-water fish farming sectors. These bacteria are described as intracellular Gram-negative coccobacilli causing high numbers of mortalities during outbreaks in the cod sector in Norway (Olsen et al., 2006). This was due to Francisella noatunensis (Mikalsen and Colqhoun, 2010). Another species, Francisella asiatica, is the aetiological agent of warm-water francisellosis affecting farmed tilapia (Jeffrey et al., 2010).

Previously, only two species of Edwardsiella were thought to cause disease in farmed fish species but comparative phylogenetic studies performed on E. tarda isolates have identified a new species called E. piscicida (Abayneh et al., 2013). Outbreaks of disease from E. piscicida in farmed whitefish (Coregonus lavaretus) were reported recently in Finland (Shafiei et al., 2016). In the last 10 years, the rainbow trout sector has suffered from emerging diseases with a suspected bacterial aetiology, including red mark syndrome (Ferguson et al., 2006; Metselaar et al., 2010) and rainbow trout gastroenteritis (Del-Pozo et al., 2010). It is hoped that further analysis will assist in clarifying the role of bacteria recovered from these infections but both continue to cause significant economic losses for the farmed trout sector.

Climate Change and Aquatic Bacterial Disease

Descriptions of climate change impacting aquaculture are often separated into direct, for example water availability, water temperature, extreme climatic events, and indirect challenges, such as transport costs, aquafeed production and costs. Of these challenges, the most significant impact on aquatic bacterial infections is temperature. All bacterial species thrive at their optimal growth temperature. Most will survive within a temperature tolerance range, where their growth may be compromised but they remain viable. Sometimes we use thermal ranges to describe bacterial pathogens as cold water (psychotrophic) or warm water (mesophiles), but the reality is often that aquatic bacteria may survive in quite large thermal ranges.

It is not unusual to see the terms pathogen/pathogenicity/virulence used interchangeably when describing a bacterial disease and yet they have very specific meanings. So, before moving on to discuss the effect of temperature on bacterial disease, let's first agree on what these terms mean. The pathogen is the organisms or bacteria which are able to cause disease in a susceptible host species under the right conditions. Pathogenicity is the ability of the bacterium to cause disease and virulence is a way to measure how it might cause disease. We might think about virulence as the tools that the pathogen has to allow it to cause the disease. Specific bacterial pathogens possess a wide range of virulence tools (sometimes called factors) which may be intrinsic to the bacterial species and found on the chromosome, such as the presence of capsules and production of endotoxins. Or they may be acquired through mobile genetic elements such as plasmids and bacteriophages. The production and secretion of the virulence factors allow the pathogen to attach, survive, reproduce and colonize the host, thus causing an infection. As we will see, aquatic pathogenicity and infectivity are directly influenced by environmental temperatures.

Climate change resulting in elevated water temperatures will alter the incidence and prevalence of the bacterial diseases currently present in our global aquaculture systems. As a direct result of rising water temperature, we may see changes in the seasonality patterns associated with some bacterial infections, such as furunculosis due to Aeromonas salmonicida or bacterial kidney disease (BKD) from Renibacterium salmoninarum. These bacterial infections often occur during rising water temperatures (Gubbins et al., 2013) as altered water temperatures will affect their incidence and prevalence.

Another outcome of climate change is the spread of bacterial pathogens into new geographical areas, resulting in emerging infections. Marcos-Lopez et al. (2010) reported that the Gram-positive Lactococcus garvieae may spread northwards within Europe due to climate change, thus directly changing the disease occurrence from this organism within aquaculture. The reported increase in Vibrio-related diseases is thought to be due to elevated sea surface temperatures again driven by climate change (Harvell et al., 2002). Vibrio species are dominant in the marine ecosystem where they can cause disease in molluscs (Paillard et al., 2004) and fish (Austin and Austin, 2016). Baker-Austin et al. (2013) correlated warming water temperatures with the increased emergence of Vibrio disease outbreaks observed in northern Europe. Such changes in water-borne infections are perhaps more of an immediate threat for shellfish farms as these production systems are located in coastal regions and already suffer from economically devastating disease outbreaks, commonly due to Vibrio vulnificus, V. anguillarum, V. tapetis and V. splendidus (Rowley et al., 2014). While water temperature will certainly affect the growth and physiological response of the host, including immune activity (Langston et al., 2002; Le Morvan et al., 1998), it will also impact on the pathogenicity of the bacterium.

Ishiguro et al. (1981) demonstrated that incubating the bacterial species Aeromonas salmonicida at higher than optimal temperatures produced spontaneous mutants with lower virulence compared with the parent strains. There are several examples of thermal restriction influencing virulence expression in aquatic pathogens. Disease outbreaks in rainbow trout fry due to Flavobacterium psychrophilum are often reported when the water temperature is between 12 °C and 14 °C (Austin and Austin, 2012) where the lower temperature of 12 °C preferentially promoted upregulation of the gfp reported gene (Gomez et al., 2012). This gene regulates proteolytic activity expressed by the bacterium, causing the degradation of the host tissues as seen in the typical presentation of external lesions (Ostland et al., 2000) during disease outbreaks. Perhaps more work is required to determine the thermosensing properties of aquatic bacterial pathogens and their influence on virulence expression. Such data would significantly improve our understanding of bacterial disease dynamics in our aquatic farming systems.

Polymicrobial and Concurrent Infections

The concept of concurrent, simultaneous and polymicrobial infections in terrestrial animal farming practice is well established and yet this is only recently being considered in aquatic disease investigation. Given that bacteria and other microbes are ubiquitous in our farming waters and our knowledge of host–pathogen interactions is still developing, it is fair to say that this is an area of future research for aquatic bacterial diseases, particularly as the trend for intensive monoculture continues.

Experimental studies have explored the impact of polymicrobial outbreaks leading to increased mortalities in aquaculture systems. Parasitic infection in rainbow trout with Myxobolus cerebralis was found to impair the fishes' immune response, leaving them susceptible to infection with the bacterium Yersinia ruckeri (Densmore et al., 2004). Increased mortality was reported in farmed Atlantic salmon that were exposed to sea lice (Caligus rogercresseyi) and then the bacterium Piscirickettsia salmonis (Lhorente et al., 2014). Phuoc et al. (2008) showed that previous exposure to the viral pathogen resulting in the disease white spot syndrome, not only impairing the shrimp immune response but also increasing the growth and establishment of a subsequent Vibrio infection in shrimp. Further experimental studies have shown that co-infections can alter the pathogenicity of bacteria, resulting in higher mortality rates (Oh et al., 2008) than those observed for single microbial infections (Dong et al., 2015).

There is therefore a need to develop more robust experimental infectivity models which include polymicrobial and concurrent infections as this reflects more accurately the real-world situation in aquaculture sites. These new experimental models should also include the role of thermal preference in disease outbreaks. This would directly support the development of novel disease control strategies and reduce production losses.

Public Health and Aquaculture

The expansion of global aquaculture will