Listeria monocytogenes, an environmental bacterium, is the causative agent of food-borne listeriosis. The clinical features of listeriosis include septicemia, meningitis, abortion, perinatal infections and gastroenteritis. L. monocytogenes infects a broad range of vertebrates including mammals, birds and amphibians. Interestingly, not all strains of L. monocytogenes seem to be equally capable of causing disease in humans, as serovar 4b isolates are responsible of all major food-borne outbreaks of listeriosis, and of the majority of sporadic cases. L. ivanovii, the second pathogenic Listeria species has a host-specificity for ruminants.
Evolution within the genus Listeria was tracked by different means to understand the specificity of strains causing disease. After sequencing the genome of the widely used strain EGDe and of an isolate of the closely related non-pathogenic species L. innocua we have determined the genome sequence of a L. monocytogenes serovar 4b strain responsible for an epidemic event in France and of a L. ivanovii strain responsible for several cases of abortion in a yew herd in Spain. Using comparative genomics of these complete sequences as well as gene content comparisons using DNA arrays and analysis of the genome sequences of isolates of each of the five species present in the genus Listeria, we gained new insight in the evolution of virulence in L. monocytogenes and L. ivanovvii.

Ecogenetics refers to gene–environment interactions, in other words, a field that studies the role of genetic variation in response to variable environments. It should be first noted that the gene does not interact directly with the environment (or in rare cases only), but through a series of fractal vehicles that contain the genes and that display phenotypes. When the word ‘ecogenetics’ is applied to bacterial pathogens, or to members of microbiotic ensembles, the main variation in the environment is expected to be produced by changes in the spatial, nutritional, immunological, and microbiotic environmental landscape, including migration into new hosts of the same or a different species. These changes are frequently secondary to the changes in the host-environment (from diseases to food intake, or malnutrition) or provoked by therapeutic action (antimicrobial agents). Indeed there is a known canalization process in most organisms, which tolerates changes (mostly cyclic or expected changes) without the need for genetic variation (phenotypic plasticity, tolerance to change), as a consequence of regulatory of physiological changes. Another possibility of adaptation to change without genetic novelty is a collective one: a number of different variants of the same organism with different habitat preferences (ecotypes) normally co-occur in a complex environment or in a set of fused environments. In the case of change, some of the variants (those more adapted to the new conditions) are enriched, but those that are not disappear, and might be re-established when the original situation returns.
In all of these examples, the gene pool might change, but without significant genetic variation. Horizontal gene transfer by mobile genetic elements (MGEs) also provides such a conservative effect, for instance when an adaptive plasmid circulates among related bacterial clones. Nevertheless, horizontal gene transfer might also produce significant genetic changes in the MGE´s gene order, because of frequent exchanges of genetic modules. An important topic is whether such a trade-off results in different MGE-ecotypes being able to adapt recipients to particular environments. The degree of modularization, both at the genetic and metabolic networks level, is increased with variable environments. Circulation of MGEs contributes to produce changes in the genetic landscape of bacterial populations, which are facilitated or prevented to follow particular cumulative evolutionary pathways. In terms of gene–protein evolution, the simplest way of detecting mutational adaptive changes in response to environmental changes is altering the environment (for instance challenging with antibiotics) and reproducing the observed changes by directed mutagenesis. An example is provided for CTX-M enzymes, in which gene–protein evolutionary trajectories can be identified using a combination of theoretical phylogenetic reconstruction and directed mutagenesis. Certainly, an integrated view of ecogenetic complexity will need advanced tools based on complex trans-hierarchical networks (systems microbiology), including ways of measuring multi-level selection. There are a lot of things to be done!

In response to the massive selection pressure exerted by the use, and abuse, of antibiotics, bacteria have become multidrug resistant. Molecular analysis of such clinical isolates indicates that, in order to do so, the strains have combined a large variety of defence mechanisms, both at the genetic and biochemical levels. A good example of this approach is provided by enterococci that have become highly resistant to glycopeptides, vancomycin and teicoplanin. Study of these strains has shown that glycopeptide resistance can only be achieved by the combination of two mechanisms: target modification associated with target elimination. In order to achieve even higher levels or broader resistance, clinical isolates have developed combinations of mutations in the acquired regulatory van genes and in resident structural genes. Dissemination of resistance among enterococci and, more recently, from Enterococcus to methicillin resistant Staphylococcus aureus (MRSA) is achieved by a two-step mechanism that combines suicidal conjugation with replicative transposition, leading to efficient transfer, stabilization, and expression of the incoming resistance genes into the new host. The extraordinary genetic flexibility of human bacterial pathogens accounts for the remarkably successful dissemination of this type of resistance that rapidly became a major public health problem.

Pathogenic Escherichia coli cause much morbidity and mortality worldwide. Two types of E. coli (enteropathogenic E. coli (EPEC) and enterohemorrhagic E. coli (EHEC, or O157)) cause severe diarrhoea, with EHEC also causing haemolytic uraemic syndrome in a subset of cases. These pathogens subvert host epithelial cells, exploiting host processes to build a cellular protrusion (pedestal) on the cell surface upon which they sit. Pedestal formation requires host signal transduction activation and major actin cytoskeletal rearrangements which are mediated by bacterial proteins that are inserted into the host cell via a bacterial type III secretion system. The host response also plays a critical role in disease development, and we have been exploiting a murine infection model to study host contribution to disease. We have also developed a bovine O157 vaccine to prevent food and water contamination. Finally we have recently begun to examine the impact of diarrheal disease on the normal microbiota during infection. Collectively, these three components contribute to disease, and only by studying all three does one fully understand the molecular complexity to bacterial disease, and use this knowledge to develop new preventatives and therapeutics.

Prior to 1998, outbreaks of bluetongue occurred occasionally in Mediterranean Europe, although these were relatively short lived (3–4 years) and involved a single virus-strain/type on each occasion. However, from 1998 onwards fundamental changes have occurred in the epidemiology of bluetongue, both in the Mediterranean region and in the rest of Europe, that have been linked to climate change. Introductions of multiple Bluetongue virus-strains (BTV types 1, 2, 4, 9 and 16) caused outbreaks, starting in Greece in 1998 (from Turkey), before spreading (with further introductions from North Africa) across the whole of southern Europe. These events are linked to the colonisation of northern Mediterranean coastal regions by the African biting midge (Culicoides imicola), which acts as a vector for Bluetongue virus (BTV). However, other midge species belonging to the C. pulicaris and C. obsoletus groups (which are abundant across Europe) have also became involved in BTV transmission in the region. Northern Europe experienced its first outbreak of BT, caused by BTV-8 from sub-Saharan Africa, starting in The Netherlands during August 2006 when temperatures were six degrees higher than previously recorded. Although it is uncertain how BTV-8 was introduced, it subsequently spread across the whole of Europe, causing the largest single outbreak on record.
During 2008 the UK vaccinated over 10 million animals against BTV-8 and was alone amongst the infected countries in successfully suppressing the disease, a landmark achievement for veterinary medicine. However, additional virus types have arrived in northern Europe, including BTV-1 and vaccine strains of both 6 and BTV-11, representing further threats for 2009 and beyond.
These events highlight the threat posed by the increasing spread of certain arboviruses (many of which can also affect humans) as a consequence of climate change.

PQuantitative risk assessment methods are increasingly used to guide standards for safe food and drinking water. The basic assumption in risk assessment is that exposure implies a risk of becoming infected and ill, dependent on the ingested dose. Epidemiological estimates of the incidence of food and water-borne disease cannot be easily reconciled with risk estimates because the two approaches measure risk on different scales. Epidemiology deals with cases of illness or death whereas the endpoint of risk assessment often is infection, including asymptomatic cases.
Novel, alternative methods for measuring infection pressure use serology as a dynamic biomarker for infection. Serology-based estimates of incidence measure the frequency of seroconversion, including asymptomatic infections. For some diseases, seroincidence tends to be higher than the incidence of notified cases of illness by several orders of magnitude, indicating substantial asymptomatic carriage.
Dose–response studies, either based on human challenge or on analysis of outbreak data, show that most human pathogens are highly infectious. The high infectivity of enteric pathogens, close to the theoretical limit where a single infectious particle has a high probability of causing infection, may be an evolutionary adaptation to environmental transmission. The conditional probability of becoming ill when infected appears to often also depend on dose: at high dose the fraction becoming ill is higher than at low dose.
The implications of both these issues: asymptomatic spread and variable attack rates, for the epidemiology of food and waterborne illness, will be discussed. The importance of reconciling risk assessment and epidemiology will also be addressed.

Early detection and fast control are two of the primary aims in preventive human and veterinary medicine. However, fast movement of people, animals and animal products in our global world makes it difficult to keep away from the incursion and spread of diseases, particularly with the current situation of global warming. This complexity is even greater when the disease of concern is a novel disease or an emerging disease, in which epidemiology, susceptible species and transmission patterns are sometimes barely known.
Epidemiological and risk-based methods has been shown to provide useful information to better detect and manage disease. Among these methods it is valuable to note in particular risk assessment, modelling, social network analysis and digital simulations. Risk assessment has become commonly used to determine and manage the risk associated with animal and animal-product transactions. Modelling techniques are particularly useful for studying the epidemiology of diseases and evaluating the cost-effectiveness of several control strategies to be applied. The recent application of social network analysis to human and veterinary epidemiology has also provided a much better understanding of the transmission patterns of disease. Finally, digital simulations provide health authorities with continual training which aim to increase the speed of the alert response and control.
This keynote presents some of the most relevant applications of risk assessment, modelling, social network analysis and digital simulations in Spain. Shortcomings, challenges and ideas for future research are also discussed.