Is a naive immune system equally able to handle new antigens as an educated one?

This is a variation of the "does the immune system run out of memory" question.

Here's a (possibly imperfect) thought experiment: You take two twins. One of them lives in a bubble from birth. One of them is Mike Rowe from Dirty Jobs, and gets exposed to all sorts of antigens as well as vaccines. When they are 30 years old, you expose both of them to an antigen that neither has seen before. What happens?

PS: Note that while this example is illustrative, my question is really about whether there is any way at all in which the response of a naive immune system to a novel antigen is different from that of a very educated immune system (which has also not seen the antigen before). For instance, would the same amount of antigen be required to stimulate the immune system? Would the antibody count several months or years post-exposure differ?

They will both go through a primary and secondary immune response. So first low affinity antibodies will bind, the corresponding b cells will undergo affinity maturation, somatic hypermutation until high affinity (perfectly fitting) antibodies are available and these antibodies will finally undergo class switching.

Depending on the antigen an innate immune response will also be triggered as part of the first line of defense. There is no difference in the immune response between these two hypothetical brothers.

Why gut bacteria are essential for a healthy immune system

The intestinal cells of the colon is covered by a protective mucus layer, which some bacteria use as food. Under the microscope, this mucus layer look like small flowers.

Most people are aware of how important it is for our well-being to have a healthy gut, which depends on a healthy gut microbiota. In fact, few things disturb our daily routines, social events or even travel experiences as the worry, pain and embarrassment of a malfunctioning intestinal system.

We even have sayings that describe how the gut can affect us: We often use our "gut feeling" to make difficult decisions, and when we are nervous of a job interview or a big examination, we have "butterflies" in our stomachs and may need to make a sudden dash to the bathroom.

Researchers are increasingly discovering and recognizing that other organ systems are influenced by the gut environment, and these links are gaining attention as possible factors in a number of diseases, such as depression and lung disease.

We may only just be beginning to discover the many ways in which a healthy or unhealthy gut can impact our lives, but we already know a lot about the important little bacteria, namely about how they impact our immune system.

1. Bacteria teach our immune system how to behave

The immune system is the main link between our gut bacteria and their influence on our health and disease. And we now know that this education begins even before we are born.

It was previously assumed that the prenatal environment in the womb was free from bacteria, but thanks to increasingly sophisticated analytical methods, we now know that bacteria are already present in the placenta. We are born with a naïve immune system and are at first protected by antibodies from our mother. However, the immune cells need to be educated further in order to learn how to protect the body from harm when the maternal antibodies are gone. This education is essential for our future health.

Bacteria educate our immune system from the moment we are born

We also know how important bacteria are for maintaining a normal immune system from experiments with germ free laboratory mice born without any bacteria at all.

These mice have an immature immune system lacking important types of immune cells. But when they are provided with even a restricted bacterial flora, the immune system matures and develops more diverse cells. These experiments have provided extensive knowledge on the function of the immune system, and of the effects of single bacteria or specific groups of bacteria.

Research conducted in both animals and humans has helped us to understand the early life factors of disease development. For example, we know that children born from caesarean section have a higher risk of developing certain diseases – some studies show as much as 20 percent higher risk of type 1 diabetes, asthma, and an increased risk of obesity, compared to vaginally born children.

This is probably due to the cleaner method of delivery, which delays the colonisation of gut bacteria and the education of the immune system. It is also known that extensive treatment with certain antibiotics at a young age increases the risk of allergy and asthma. The hygiene hypothesis has led the way to this line of thinking, but other factors such as antibiotics use of the mother and pre-term planning of caesarean sections with immature maternal milk may also influence these increased risks of disease.

2. Gut bacteria maintain a balanced immune system

Throughout life, we are constantly exposed to new things in our gut, nose and lungs, via our food and environment, such as food additives, pollen in the air or non-pathogenic microorganisms in dust or dirt. But thankfully, most people have healthy immune systems that handle all of these invading objects with ease.

If it didn't, it would elicit an inflammatory response every time you tried a new food or visited a new country with different types of trees. This would be a highly ineffective and unnecessary use of energy.

The essential task of the immune system is to maintain a balance between reaction and tolerance. It is essential that this tolerance, called oral tolerance, is established. And a diverse gut flora established in early life with many types of bacteria, fungi, and other microorganisms, is crucial for this, as it teaches the cells of the immune system that not everything is bad.

Since balance of bacteria in our gut influences the balance of our immune system, an unbalanced bacterial flora with for instance too many opportunistic pathogens can shift the immune system to an increased inflammatory state with a so-called "leaky gut". This inflammatory state may then affect other body systems and increase the risk of obesity, type 1 and type 2 diabetes and even depression.

3. Bad gut bacteria can lead to disease

Most bacteria are beneficial, but some are responsible for the progression of disease.

It is perhaps common sense that gut bacteria play a significant role in diseases directly related to the gut, such as inflammatory bowel diseases. This has been studied for years and today, treatments are available to correct skewed bacterial compositions and aid recovery of beneficial bacteria via faecal transplantation in some colitis patients. Most people are also familiar with the use of over-the-counter probiotics especially during exotic vacations.

Bacteria are survivors in the best Darwinian style, and they will to some extent adjust to the environment they are in. This is, for instance why resistance to antibiotics occurs. This also means that if good bacteria are removed due to for example diet or medication, some of the opportunistic commensals, or pathogens, will immediately move in and try to fill the gap.

A diverse gut flora is the healthiest

It is not so easy to permanently change an established gut flora, good or bad. Once disturbed, the flora will return to normal within a short time frame, just like when you return home after a vacation and eat your usual diet.

But an imbalanced gut is able to loop in a bad cycle, whereby harmful functions are reinforced. In laboratory mice, researchers have shown that a certain bacterial composition is associated with type 1 diabetes and obesity – in fact, researchers were able to transfer obesity to lean mice by transplantation of the gut microbiota.

Such skewed microbiotas all have one thing in common: a lack of diversity. A diverse microbiota is more likely to bounce back from unhealthy fluctuations in diet and withstand outside intruders, and this means a much more tolerant and well-regulated immune system.

The colon, here seen from within, contains more than 10,000,000,000,000 cells per gram of intestinal content and between 300 to 1000 different bacterial species.

Gut bacteria could lead to personalised Microbiota Transfer Therapy

So how can we use all of this knowledge in the future?

We know that presence or absence of bacteria is important in the development of several diseases. We also know that it is rarely just one or two bacterial strains that make a difference, but more likely a whole group of certain bacteria influencing other bacteria.

This is all very challenging to study in humans—especially in complicated scenarios, where these skewed bacterial communities cause trouble elsewhere in the body.

Until now, scientists have focused on understanding the presence or absence of certain bacteria, but what really interests us today, is what these bacteria produce and what signals they send to the rest of the body. Luckily we now have advanced tools at our fingertips to figure this out.

Systems biology with whole genomic, whole proteomic, and whole metabolomics analyses are revealing new details about these bacteria and might even lead to personalised diagnosis and treatment. For instance, it is likely that in the near future, the examination of patients will include a full assessment of the microbiota or its products just like a routine blood sample, leading to precise interventions in diet or administration of bacteria.

Let your kids get dirty

In addition, these methods help explain other mechanisms in the body related to bacteria. For example, a 2017 Nature paper showed that some of the beneficial effects of the type 2-diabetes medicine "Metformin" that enhance insulin sensitivity in type 2 diabetic patients, are due to its effect on gut microbiota and their products. In particular due to the promotion of the good bacteria Akkermansia Muciniphila.

Using these methods, we can establish clearer cause and effect relationships between bacteria and outcomes, which have previously been difficult. In other words, we are a step closer to tracking down exactly which part of the gut microbiome is different in a disease state, improve it with diet, medicine or bacterial transplants, and follow the change in bacterial products and messengers.

Researchers will probably soon be able to buy their laboratory mice with a "diabetes – or obesity" inducing gut microbiota or even with a humanized microbiota. This could improve our disease models and make them more effective. It might even help us understand what circumstances are necessary to really permanently change a person's gut microbiota to the better.

Research in nanotechnology is producing new ways of delivering medicine, vaccinations, and bacteria to the body. Imagine a nano-sized container with a specific bacterial mix meant for the distant part of the gut,designed to protect the bacteria and only open when they meet the appropriate "key" at the right location, for instance an enzyme or a specific pH value.

Clinical studies of microbiota transfer therapy in humans are already taking place and probiotic use is increasing (autism spectrum disorder improved with faecal therapy) and there is no doubt that new and more specialized probiotics will be presented in the near future (for example, the NxtGenProbio project is expected to yield interesting results).

Personalised bacterial "diagnosis" and treatment would certainly be a valuable tool for health professionals, but it is unlikely to become a commonly used tool any time soon since there are still many unknown factors and risks. For instance, should a faecal "donation" come from your own gut, or from a different part of the intestines? How do we prevent transfer of bad bacteria along with the good ones? Are family members more compatible donors compared to a standard foreign donor?

Until then: Let your kids get dirty with a good conscience. you are priming their gut flora into being balanced and healthy.

The many axes of gut bacteria

Signals run along axes from the gut to other parts of our bodies via neurons, hormones, and perhaps most importantly via the immune system. We call these "axes" and they help describe the connection between gut bacteria and disease else-where in the body.

The most studied axis so far is the connection between gut and brain, since it is documented and well-known among health professionals that patients suffering from inflammatory bowel diseases often also suffer from depression.

The gut is able to alter the brain chemistry via neuronal pathways and through messengers of the immune system, called cytokines – and these messengers depend on the state of the gut microbiota.

Stress is a good example: stress changes the gut microbiota, and the signals running to the brain may impact how we behave. For instance, early life stress changes the gut microbiota of monkeys, and rat pups which are stressed by separating them from their mothers prematurely. Their gut microbiota is disturbed as a result, and they have increased levels of stress hormone and a different immune response.

Another axis is the gut-liver axis, which is studied widely in liver research, since 70% of the blood flow to the liver is directly flowing from the gut.

Gut bacteria are a vital source of fat components and of circulating antigens, and may impact the risk of fatty liver disease.

3. The gut-lung and gut-kidney axes

The gut-lung axis is of interest in respiratory disease research, where the gut microbiota influences both asthma, COPD, pneumonia and even development of cancer.

Scientists have also proposed a gut-kidney axis where the bad toxic products of a diseased kidney affect the microbiota and a bad microbiota increase the amount of toxins released as a disease mechanism in chronic kidney disease.

This story is republished courtesy of ScienceNordic, the trusted source for English-language science news from the Nordic countries. Read the original story here.


We are witnessing a major change in immunology’s conceptual character from an emphasis on immunity as a defense to immunity as an interface-of-exchange. Immunity should be regarded as a communicative system of the internal homeostasis which perceives and then mediates environmental information (organic and inorganic internal and external) [1]. To handle this complexity there is an increasing need for complex computational models to perform in silico experiments as an adjunct to in vitro and in vivo experiments. One of the key points of immunity is the concept of self-nonself discrimination. We proposed first that in order to recognize self and non-self, T lymphocytes should recognize the much smaller set of self antigens, rather than the practically unlimited non-self antigen universe [2, 3]. The immune system is continuously in a state of delicate balance between tolerating self and attacking non-self. If this balance is perturbed, autoimmune reactions occur. Immunological tolerance is rooted in regulatory immune cell subsets, suppressive cytokines, and immune checkpoint pathways [4].

A good example for the delicate balance between immune tolerance and intolerance – and for the importance of this research area – is the ambiguous results of The Cancer Immunotherapy Revolution [5], in which the newly approved immunotherapies manipulate components of the immune system to attack tumors. Hundreds of clinical trials are underway to improve responses and success stories of terminal cancer patients defying the odds and achieving complete remissions are accumulating. Unfortunately, the manipulation of the immune system has also resulted in a major safety issue: the iatrogenic immune-related adverse events (IrAEs). As a result of the impaired self-tolerance, irAEs may present with a broad clinical spectrum that mainly involves the gut, skin, endocrine glands, liver, and lung but can potentially affect any tissue, and their incidence may reach up to 90% of patients [6, 7].

In order to aid in the qualitative characterization and examination of the delicate immune balance, we have developed MiStImm computer program, which is capable to simulate the complex processes of self-nonself discrimination of the adaptive immune system. We know that a computer model can not reliably simulate the whole immune system, however, simulating areas of interest can be useful for testing ideas to help in the design of in vivo and in vitro experiments [8].

MiStImm uses agent-based modeling technique [9] and it can simulate some aspects of humoral immune response along with its major components, including T cells, B cells, antibodies, danger signals, interleukins, self cells and foreign antigens. These simulation components (called “agents”) determine the nodes of a dynamic immune network where links are the potential interactions between two elements. The immune network changes step by step (in time) driven by random events. Using the terminology of [10], a model simulated by MiStimm is an agent-based model that is in part “individual particle based-stochastic”, and in part “particle number stochastic”. An “individual particle based stochastic element” is an agent that models individual cells and their individual random attachments with other cells or molecules. In our program, this approach is used for Th cells and B cells. A “particle number stochastic element” is a population of cells or molecules that are represented in the model by the properties of the population and by the number of elements in it. In our program, this approach is used e.g. for self cells and foreign antigens. Because our model is stochastic, their attachments with other elements is also controlled randomly. A great advantage of such a model is that it can easily incorporate the most important types of cells and molecules together with their essential features and simulation events that play important roles in immune reactions. In such a simulation events – for example interactions of components – occur at random. A stochastic model fits well with the affinity maturation of B lymphocytes in which random events are perhaps the most characteristic. It is also suitable to model the development of the regulatory T cell population and the random selection of specific T cell clones.

To simplify things, we chose the humoral adaptive immune system since the humoral phase (blood or lymph) may be considered spatially homogeneous thus a microscopic spatial volume may represent the whole phase well. A major advantage of this approach is that it is not necessary to describe the actual spatial positions and spatial motions in the model. Instead, model components (agents) randomly choose one of the other components as interaction partners, because any components are close enough to become engaged in an interaction.

As the first application of MiStImm, we have simulated two different immune models and then we have compared performances of them in the mean efficacy of self-nonself discrimination (see the Results). The first model is called nonself centered or Conventional Role of Self (CRS) where even a primary immune reaction depends on the recognition of non-self antigens by T and B cell receptors [11–13]. The role of self in this model is that the great majority of autoreactive T and B cell clones are selected and purged from the immune system [14]. The second model called self-centered or Enchanced Role of Self (ERS) which is based on our previously published “one-signal model” [3]. We proposed that model (hypothesis) when we have been seeking the answer to three unresolved paradoxes of immunology:

(Q1) How can a tiny fraction of human genome effectively compete with a vastly larger pool of mutating pathogen DNA [15]?

(Q2) Considering the fact, that average 3 mutations are formed each of the 10 16 times the cell’s 3·10 9 DNA base pairs are duplicated during a human lifetime [16], “why does cancer occur so infrequently”?

(Q3) Considering the facts that T cells require three to five days to attain fighting strength (because they are rare, short-lived, and their doubling time is at least 6 h), yet how can a T cell response be measurable in the lymph nodes draining the infection site within 12 to 18 h [17]?

In order to explain these paradoxes, we have suggested a new T cell model [3] that we can summarize in the following. We have postulated that a dynamic steady state, a so-called coupled system is formed through low affinity complementary TCR–MHC interactions between T cells and host cells. Under such condition, it is sufficient to recognize what is self in order to attack nonself (answer to Q1). We have postulated that the evolutionary pressure driving the creation of the T cell receptor (TCR) repertoire was primarily the homeostatic surveillance of the genome (answer to Q2). The new model implies that a significant fraction of the naive polyclonal T cells is recruited into the first line of defense from the very outset of an infection (answer to Q3). The computational variant of our hypothesized T-cell model is the ERS model, presented in this paper. The ERS and CRS model are summarized by Fig. 1.

Humoral adaptive immune response by the ERS and CRS model. The ERS model is described by (a), (b) and (c), while CRS models are described by (c) alone. a In the ERS model, a hypothesized weak affinity interaction begins in intrauterine life and keeps the immune image of self during the whole life. It is sufficient for homeostasis low affinity BCR binds self-antigens and presents self-peptides in their MHCII to regulatory T helper (Threg) cells this ensures B and Threg cell survival. b In the ERS model another hypothesized interaction, intermediate affinity interaction initiate the first line of defense against an infection some B cells that have higher BCR affinity for the antigens of the pathogen capture pathogens with intermediate affinity and present foreign peptides in their MHCII. The foreign peptides indirectly inhibit binding of Threg cells to these B cells for a critical time period, then the B cells will secrete hypothesized danger signals. Danger signals activate local Th cells, which in turn, release interleukins that fuel local T cell activation. This way a non-specific, local polyclonal B and T cell activation is induced, which is the primary defense mechanism against infections in the ERS model. Clonal expansion requires affinity maturation, which results in a several magnitude increase of BCR affinity, typically over a time of one week. Random mutations cause the production of B cells with a broad range of affinities for their presented foreign antigen. B cells with unfavorable mutations will not get sufficiently activated by the foreign antigen and will die, while those with improved affinity will be stimulated to clone themselves. c Specific immune reaction, here called as strong affinity interaction, appears in both the ERS and CRS models and is supervised and supported by pathogen peptide-specific Th cells, which require direct contact via TCR to the MHCII of the expanding B cell clone. Such higher affinity interactions would then drive clonal T cell proliferation, activation, lysis of infected cells. Having cleared the infection, specific T cells could eventually become an expanded memory type T cell clone, while B cells could differentiate into infection specific antibody-producing plasma cells or memory B cells. This interaction usually needs several days to efficiently start

Though, there are some immune system simulation models that are capable to simulate a conventional (or standard) immune model like our CRS model (e.g. Basic Immune Simulator [18], C-ImmSim [19, 20], SIMISYS [21]), they are not directly usable for modelling the self-nonself discrimination theory of our group (presented by the ERS model) and compare it with a conventional model. However, when we built MiStImm, we have adapted some principles of the earlier simulation models (see the Discussion for a comparison).

The main goal of the simulation experiments of the current paper is showing that the ERS model matches real patterns and additionally to analyze how the two models (CRS and ERS) cope against a critical primary infection. That was the main reason why we have developed MiStImm.

In the Results we show that the ERS model does not develop autoimmune reactions despite the existence of the hypothesized TCR–MHC interaction between T cells and self antigens in the model. Autoimmune reaction is a strong immune response of an organism against its own healthy cells and tissues. Despite the weak reaction of B and Th cells against the healthy self cells in the model, the sizes of these self cell populations are not decreasing, so does not occur a pathological consequence of these weak reactions.

We also show that the ERS model gives better results to overcome a critical primary infection, answering the paradox “how can a tiny fraction of human genome effectively compete with a vastly larger pool of mutating pathogen DNA?” We hope that our results will encourage investigations to make in vitro and in vivo experiments clarifying questions about self-nonself discrimination of the adaptive immune system. We also hope that MiStImm or some concept in it will be useful for implementation and/or comparing other immune models by other researchers.


Ticks (Acari: Ixodida) are ectoparasitic arthropods that obligatorily feed on the blood of a diverse list of vertebrate hosts, including mammals, birds, reptiles, and even amphibians. More than 950 tick species have been described to date, which, according to morphological and physiological characteristics, are divided into two main families, Ixodidae (hard ticks), comprising more than 75% of tick species, and Argasidae (soft ticks) a third family, known as Nuttalliellidae, is monospecific (1, 2). As a result of blood spoliation [a single ixodid adult female can ingest more than

1 mL of blood (3)], the host can suffer from anemia, which negatively impacts the productivity of livestock and causes a huge economic burden worldwide. For example, the estimated annual losses due to reductions in weight gain and milk production caused by the cattle tick Rhipicephalus microplus are approximately 3.24 billion dollars in Brazil alone (4).

In addition to ingesting blood, ticks also secrete saliva into the host during feeding. Tick saliva, produced by their salivary glands, returns excess water and ions to the host, thereby concentrating the blood meal (5). Tick saliva contains an arsenal of bioactive molecules that modulate host hemostasis and immune reactions, thus enabling blood acquisition (6, 7). The antihemostatic and immunomodulatory properties of saliva can also facilitate the infection of pathogens that use saliva as a vehicle to be transmitted to the host during tick blood feeding (6, 8). Indeed, ticks are versatile vectors of viruses, bacteria, protozoans and nematodes, which cause life-threatening diseases to humans as well as to other animals, including livestock, pets, and wildlife (9). Among human diseases, we highlight Lyme disease, the most common tick-borne zoonosis, which is caused by spirochetes from the Borrelia burgdorferi sensu lato complex. After transmission by the bite of an infected tick, the typical clinical sign of Lyme disease is erythema migrans, but infection can spread and affect joints, heart, and the nervous system (10).

The first organ that a pathogen acquired within the blood meal interacts with is the tick gut (Figure 1). Then, the pathogen must colonize the gut epithelial cells and/or cross the gut epithelium to enter the hemocoel, an open body cavity filled with hemolymph, the fluid that irrigates all the tissues and organs in the tick. The pathogen must then reach the salivary glands. In each of these organs, the pathogen must counteract tick immune factors to be successfully transmitted through saliva to the vertebrate host in a subsequent blood-feeding (11). Some pathogens also have the ability to invade tick ovaries and can therefore be transovarially transmitted to progeny (Figure 1). Thus, elucidation of the immune factors involved in the interactions between ticks and tick-borne pathogens (TBPs) in each of these steps is essential to understand the biology of tick-transmitted diseases and may help to identify targets for the development of new strategies to block pathogen transmission. In this review, we present an update on humoral and cellular tick immunity components (Figure 1), including signaling pathways, antimicrobial peptides (AMPs), redox metabolism, complement-like proteins, and regulated cell death. Using a comparative approach with the immune system of other invertebrates, we highlight the challenges of studying tick immunity, the gaps, such as prophenoloxidase (PPO) and coagulation cascades, and the interconnections, such as immune system signaling pathway crosstalk. In addition, the role of tick microbiota in vector competence is also discussed.

Figure 1 Main interactions among tick immune system components, microbiota, and pathogens. Pathogens ingested within the blood meal initially reach the tick gut, where they interact with components of the gut microbiota and with cytotoxic molecules, such as AMPs (hemocidins and endogenous AMPs) and possibly with factors of redox metabolism, despite not being fully comprised. Pathogens must colonize and/or cross the gut epithelium to reach the hemocoel, which is filled with hemolymph. In hemolymph, complement-like molecules attach to pathogens that can be engulfed or trapped by hemocyte-mediated processes named phagocytosis and nodulation, respectively. Invaders can also be killed by several types of effector molecules, including AMPs, complement-like molecules, and factors of redox metabolism. The tick salivary glands return excess water and ions from the blood meal to the host through saliva, which also contains antihemostatic and immunomodulatory molecules. Pathogens use tick saliva as a vehicle to be transmitted to the host, in which infection can be facilitated by saliva properties. Some pathogens can also colonize the tick ovaries and are transmitted to progeny. In the tick salivary glands and ovaries, as in the gut, pathogens must deal with the members of resident microbiota as well as tick immune reactions. Additional studies are required to elucidate the molecules responsible for hemolymph clotting and melanization in ticks.

Selective killing

The most desirable immune response is one that stops an infection in its tracks, before it has established a foothold in the body. Phagocytosis of bacteria in the tissues and antibody-mediated blockade of virus entry into cells work this way. But if, once an infection was established within a cell, the immune system did no more, or if a cell that had turned into a cancer was ignored, viruses and cancers would be unstoppable. To deal with these eventualities, cells of the immune system control powerful lethal weapons. This ability is so striking that the cells that specialize in execution are known as cytotoxic killer cells.

Killers discriminate by using recognition receptors. Cytotoxic T-cells use the TCR and the CD8 co-receptor which together interact with MHC I. In this way, they can interrogate any nucleated cell in the body. When cytotoxic T-cells recognize an infected target cell, they kill it rapidly and move on to the next cell.

Natural killer cells, part of the innate response, use a different approach to selecting their targets. These cells patrol the body asking themselves whether the tissues that they survey express MHC I molecules. If the cells that are being examined do, then they move on, but if not, they will become activated and kill. This provides an alternative method of surveillance, which does not depend on specific antigen, and frustrates a strategy that some pathogens employ, of inhibiting the surface expression of MHC I. By insisting that nucleated cells report on their protein production, the window of opportunity that viruses have to squeeze through to be successful is narrowed further.

Cell killing is a specialized function that occurs in a series of steps. First the cytotoxic cells make close contact with the target and mobilizes intracellular granules to this area of contact. Then these granules fuse with the cell membrane of the cytotoxic cell and release a number of proteins. One, called perforin, forms a pore in the cell membrane of the target. This pore allows the entry of other proteins called granzymes that trigger rapid cell death. A second pathway, that probably plays a minor role in cytotoxicity, is that involving a receptor on the cytotoxic cells called FasL binding to its partner on the target cell called Fas. This interaction triggers a suicide signal within the target cell.

All cells are able to commit suicide and this type of cell death is called apoptosis it is very important in development and in the immune system. In the thymus, where, as explained above, most of the lymphocytes that audition for a role as useful effector cells die, it is through the process of apoptosis that this occurs. Apoptosis also frustrates some viral infections and protects individual cells from becoming cancerous. Cancers only develop if they have mutations that block the activation of apoptosis, and viruses produce proteins to stop apoptosis switching on. In this way, the transformed or infected cell can survive signals that would otherwise lead to it killing itself.

The final common pathway of apoptosis is a proteolytic cascade which digests the contents of the cell and fragments its genetic material the cell shrivels up and exposes signals at its surface that tell neighbouring phagocytes to eat it. The enzymes that carry out this process come from a family called caspases (cysteine proteases that cleave proteins after aspartic acid residues). There are several different ways to initiate this process, but they share a common final pathway. One feature of ‘normal’ apoptosis, such as occurs in the thymus, is that it does little to stimulate inflammation. This means that cells can die without initiating an effector immune response. In contrast, in an aggressive infection, where cells death occurs alongside signals that stimulate innate immune activation, accompanying adaptive responses will also occur. Successful immune responses reach an appropriate match between the threat and the response, producing enough killing to manage infection, but not so much that the host is compromised. If this is not achieved, the outcome may be disastrous in the short-term, because an infection is not controlled, or because the immune response is so aggressive that the body collapses. Or dangerous in the long-term because a chronic infection becomes established or because the immune system over-reacts and develops specific responses that attack healthy tissue.

Materials and Methods

The Computational Model


In the present computational model, the specific recognition in adaptive immunity is simulated by borrowing ideas from binary calculus (14). Epitopes and paratopes are represented by strings of zeros and ones. When an epitope meets a paratope the strings are checked for complementarity at each position and a match (or equivalently a mismatch) is scored. Thus, the match is a number between 0 and N where N is the length of the binary strings representing the two binding regions. The model is polyclonal since it equips cells and molecules (e.g., lymphocytes receptors, B-cell receptors, T-cell receptors, Major Histocompatibility Complexes (MHC), antigen peptides and epitopes, immuno-complexes, etc.) with specific bit strings to represent the 𠇋inding site.”

This minimalistic definition allows a diversity of 2 N for each immunocyte (CD4+ or Th, CD8+ or TC, B). Such a setup can model cross-reactivity with remarkable smoothness, and accuracy in predicting the effect of competition among cross reactive cells.

Binding Affinity

In vivo, the paratope-epitope attraction is the sum of weak electrostatic and hydrophobic interactions when juxtaposed. In the simulation, two entities interact with a probability that is a function of the Hamming distance between the binary strings representing the entities' binding site. We indicate with m = ||r, p|| ∈ <0 … N> the distance or the match between r, p ∈ <0 … 2 N − 1>. A good and widely used analogy is the matching between a lock and its key. If more than a threshold value mc over N bits matches (i.e., 0𠄱 or 1𠄰) occur, the interaction is allowed with a certain probability that is a function of the number of matches between the bit-strings. This attraction force (called affinity or affinity potential) is equal to one when all corresponding bits are complementary. Specifically, if m = ||r, p|| is the Hamming distance between the two strings r and p, the affinity potential f (m) ∈ [0, 1] defined in the range 0, …, N is

where AL is a free parameter which determines the slope of the function, whereas mc ∈ <N/2 … N> is the cut-off (or threshold) value below which no binding is allowed.

Humoral and Cellular Responses

The model simulates a very simple form of innate immunity and an elaborate form of adaptive immunity (including both humoral and cytotoxic immune responses).

In the case of innate immune response by 𠇎xogenous signal” (e.g., Pathogen-Associated Molecular Pattern, PAMP or PAMP-agonist, used for specific adjuvants) the activation sequence will begin with antigen presenting cells stimulation. The only mechanisms of this kind which is embedded in the model accounts for the presence of lipopolysaccharides in pathogens as in Gram-negative bacteria.

Working Assumptions

In the model, a single human lymph node (or a portion of it) is mapped onto a three-dimensional Cartesian lattice. The primary lymphoid organs thymus and bone marrow are modeled apart: the thymus (15, 16) is implicitly represented by the positive and negative selection of immature thymocytes before they enter the lymphatic system, while the bone marrow generates already mature B lymphocytes. Hence, only immunocompetent lymphocytes are modeled on the lattice.

The C-IMMSIM model incorporates several working assumptions or theories, most of which are regarded as established immunological mechanisms, including: (i) the clonal selection theory of Burnet (17) (ii) the clonal deletion theory (i.e., thymus education of T lymphocytes) (18) (iii) the hypermutation of antibodies (19) (iv) the replicative senescence of T-cells, or the Hayflick limit (i.e., a limit on the number of cell divisions) (20) (v) T-cell anergy (21) and Ag-dose induced tolerance in B-cells (22) (vi) the danger theory (23) (vii) the idiotypic network theory (24). Variations on the basic model have been used to simulate different phenomena ranging from viral infection [e.g., Human Immunodeficiency Virus (25) or Epstein-Barr Virus (26)] to cancer immunoprevention and type I hypersensitivity (27, 28).

Each time step of the simulation corresponds to 8 h. The interactions among the cells determine their functional behavior. Interactions are coded as probabilistic rules defining the transition of each cell entity from one state to another. Each interaction requires cell entities to be in a specific state choosing from a set of possible states (e.g., naïve, active, resting, duplicating) that is dependent on the cell type. Once this condition is fulfilled, the interaction probability is the effective level of binding between ligand and receptor.

Unlike many other immunological models, the present one not only simulates the cellular level of the inter-cellular interactions but also the intra-cellular processes of antigen uptake and presentation. Both the cytosolic and endocytic pathways are modeled. In the model, endogenous antigen is fragmented and combined with MHC class I molecules for presentation on the cell surface to CTLs' receptors, whereas the exogenous antigen is degraded into smaller parts (i.e., peptides), which are then bound to MHC class II molecules for presentation to the T helpers' receptors.


The stochastic execution of the algorithmic rules, as in a Monte Carlo method, produces a logical causal/effect sequence of events culminating in the immune response and development of immunological memory. The starting point of this series of events is the injection of antigen (the priming). This may take place any time after the simulation starts. In general, the system is designed to maintain a steady state of the global population of cells if no infection is applied (homeostasis). Initially the system is “naïve” in the sense that there are neither T and B memory cells nor plasma cells and antibodies. The various steps of the simulated immune response depend on what is injected, i.e., virus or bacteria.

The Virus

Virus is the 𠇏oreign agent” in the model. It is constructed with B-cell epitopes and T-cell peptides. In addition, it replicates, simulating a living entity, and the combination of three factors (speed of duplication, infectivity, and lethal load level) results in its 𠇏itness” which is independent of antigenicity. Any infection begins with the penetration of virus into an epithelial cell, though this could be any designated target cell. Whether the infection is cured or becomes persistent or even kills the virtual mouse depends on the virus dose, its fitness, and the strength of the immune response it has elicited. All these variables determine whether𠅊nd to what degree—the immune system's success requires the cooperation of both the cellular and humoral branch, as has been shown in several simulation studies (13).

Modeling Active Attrition

Active attrition is enacted in the present version of the model by describing the release of IFN-β by macrophages in the presence of high concentrations of danger signals, e.g., in infection sites. This lymphokine diffuses locally and then �uses” the death of cytotoxic memory T-cells by contact. The locally-limited bystander effect of this cytokine is set to be dependent on the cell's age but also on its affinity to the viral peptide. Specifically, the death of cytotoxic cells is modeled as a stochastic event whose probability is proportional to the cell's age and inversely proportional to the affinity between TCR and the peptide attached to class 1 HLA (1, 5) of infected cells, i.e.,

where a is the age of the T-cell (in units of days), f the affinity of its TCR to the viral peptide as defined in Equation (1) and i is the local concentration of IFN-β (in pg/ml). In the experimental setup that we are going to describe in the following section, the parameters of Equation (2) have been chosen as follows: k 1 = 1 0 6 × days - 1 and k 2 = 1 0 9 × ( pg / ml ) - 1   w e r e taken to obtain a probability of killing which was much stronger for memory compared to naïve cells parameter n1 = 3 > n2 = 2 were chosen to make age the limiting factor in the killing. The last term in Equation (2), (1 − f) ∈ [0, 1], stands for a protective factor for cells able to establish a stronger immunological synapse during peptide recognition on the membrane of infected cells and f therefore is the same function in Equation (1).

Experimental Setup

The model represents both paratopes and epitopes by N = 16 bit binary strings. A successful paratope-epitope interaction is limited to a match m greater than or equal to the cut off mc = 13 over the 16 allowed. This setup results in a diversity of 2 16 for each lymphocyte and gives Nmc = 4 matching classes thus allowing to model the immune recognition and predicting the effect of competition among cross-reactive cells with reasonable accuracy. In vivo, the diversity among epitopes and that among paratope are mind boggling (conservatively, 10 10 to 10 12 ). Simulating those numbers, though theoretically possible by enlarging the repertoires which is obtained by elongating the strings, is practically not viable for computational reasons.

The Antigenic Distance Experiments

In studying memory, it is important to quantify the degree of cross-reactivity between related antigens. While in vivo this appraisal is difficult to attain, the following modeling setup allows us to measure the effect of cross-reactivity on a secondary immune response quite effectively.

The series of simulations we perform mimic a prime/challenge experiment in a virtual mouse (or individual) where successive injections carry equal or different antigenic determinants (see Supplementary Figure 2). The priming infection is performed always with the same virus, but the challenge or secondary infection performed later is done with a different virus whose peptide is at a defined distance d from the priming one. We use N/2 = 8 bits to represent a virus peptide thus we have d = 0 … 8 levels of cross-reaction by suitably choosing the prime/challenge couple. Viral peptides are presented to T-cell receptors bound to the major histocompatibility complex molecule (MHC) and indeed in the model the match is an N-bit match. However, for simplicity, the contribution to the affinity given by the portion of the cell receptor binding the portion of the MHC molecule is set to a constant value so not to influence the overall match to the virus. In other words, the affinity between receptors and MHC-bearing virus peptides depends only on a N/2 = 8 bit match rather than an N bit match.

Let's call V I the virus injected first (i.e., the primer at time tI), V II the virus injected subsequently (the challenger at time tII) and d the 𠇋it distance” between V I and V II , that is, d = ||V I , V II ||. The experiments realize the protocol consisting in a priming injection that is always performed with the same virus V I = V 0 and a challenge injection consisting of a certain saturating dose of one of the nine viruses reported in Table 1 which also includes V0. Therefore V II = V k for k = 0 … 8. Note that the set of chosen viral peptides is such that d = ||Vi, Vj|| = |ij|, for all choices of i, j ∈ <0 … 8>. Following this description, it is convenient to name the experiments on the basis of the distance between priming with V0 and challenge with Vk. For instance, we call d = 3 the experiment in which V I = V 0 and V II = V 3 because d = ‖ V I , V II ‖ = ‖ V 0 , V 3 ‖ = 3 . While d = 0 realizes the homologous response, and can indeed be considered the control, d = 1 to d = 6 represent cases of cross-reactivity, with progressively fewer matches. Finally, d = 7 and d = 8 are heterologous responses (i.e., no match at all). We note that all viral peptides are chosen to be distant with respect to self-peptides, to avoid having to deal with autoimmune responses, which are outside the scope of this work.

Table 1. Viruses used in the experiments are numbered from zero to eight.

The simulated space is equivalent to a fraction of the lymphatic system represented at once. This simulated volume is 10 micro liters or, equivalently, 10 cubic millimeters. Both priming and challenge consist in injecting a saturating viral dosage of 10 3 viral particles per microliter. For all experiments, the setup is identical except for the two viruses injected, V I and V II . Thus, the simulated space is populated with the same initial number of cells (i.e., no variability allowed), the viruses share the same infectivity and replication rates, etc. Moreover, since the model is stochastic, for each setup d ∈ <0 … 8>, we repeat the experiments 100 times for each protocol and calculate statistics (averages and standard deviations) afterwards.

Useful Definitions

With the aim of defining two quantities which help in measuring the effect of cross-reactivity, we now need to introduce some formalism.

We call diversity D the set of possible bit strings of length N in the base-ten system, that is, D = <0𠉢 N/2 − 1>. We indicate by nr(t) the number of cytotoxic T-cells with specificity rD at time t. For each virus V the Hamming distance creates the equivalent classes in the set of cell receptors D. In other words, two receptors r1 and r2 are in the same matching class for V if ||r1, V|| = ||r2, V|| = m. We can therefore define qm(t) as the total number of cells matching the virus V with m bits, that is, ∀m ∈ <0 … N>

Then we call Am(t) the affinity of the response to the peptide of virus V relative to the matching class m ∈ <0 … N>, that is, all the lymphocytes that are equivalent in terms of affinity. This quantity is calculated by summing the number of cells with receptor matching with m bits the virus peptide and multiplied by the affinity value f(m), that is, ∀m ∈ <0 … N>

Finally, we define the total affinity to virus V as

Note that since we are interested in quantifying the effects of cross-reactivity on the secondary immune response, all the quantities qm(t), Am(t) and TA(t) should be considered relative to V II and be written, for instance, A m I I . However, to simplify the representation we just avoid using the superscripts and write Am, etc.

Furthermore, we call V II (t) the number of viral particles at time t of the challenge virus and define

the time-to-clear the virus V II , that is, a measure of how quickly the response eradicates the virus injected at time t II .

We can now finally define two almost complementary measures. The first one is the efficacy of the immune response to the second virus injected V II . The efficacy E(d) is a function of the distance

to the first injected virus V I = V 0 and is defined as the peak value of TA(t) for ttII divided by the time-to-clear the virus tc. In formula

The efficacy measures how good the immune response to V II is in terms of how many cytotoxic T-cells are developed by clone expansion and how quickly the virus is eliminated. Clearly the maximum value of the efficacy is achieved for d = 0 because of the immune memory developed to respond to V I = V 0 , but decreases for increased distance d between prime V0 and challenge injection V II .

The maximum value attained by the sum of all Tc counts qm(t) for m = 0 … N averaged over a number of simulations (〈·〉 indicates averages) can be designated as

Cell counts are calculated for each antigenic distance experiments. We can therefore use superscripts to indicate a specific experiment and refer to this quantity in the case d = 8 as 〈 M ~ 〉 d = 8 . This value measures the magnitude of the cytotoxic immune response to V I I = V 8 . Since it corresponds to the completely heterologous response, the effect of the MaN is zero and the quantity in Equation (6) is maximal with respect to d. The other extreme case is found when d = 0, corresponding to a homologous immune response for which the immune memory is so perfectly fit to the second injected virus V I I = V 0 = V I that the latter is eliminated without the need for a clonal expansion of cytotoxic T-cells. The measure that we call compression is then defined as

and is the difference of the maximum number of cytotoxic T-cell count attainable in the absence of memory. In other words, this measure quantifies the degree of hindrance (or reduction, hence the name compression) of the naïve response due to the presence of cross-reactive memory cells against past infections. The compression is maximal for d = 0 and diminishes for larger d reaching its minimum for d = 8.

In vivo studies

The most physiologically relevant model of human immunity is the study of humans themselves in health and disease. Understanding immune variation among people can also tell us a great deal about how the immune system functions as a holistic unit during steady state and immune perturbations. Experiments dating back to just after the 1918 influenza pandemic indicate that people volunteered for infection challenge studies to improve understanding of disease transmission, immune memory, and the clinical course of infection [140,141,142]. Current human in vivo studies undergo rigorous ethics review and, for human challenge models in particular, health checkups prior to participation are part of the inclusion/exclusion assessment [143]. In vivo studies can tell us about the fundamental nature of immune cell functions, such as homeostatic proliferation and memory retention, that previously were almost exclusively studied in mice. For example, in a recent 10-year study of yellow fever vaccine recipients, Akondy et al. [144] determined that long-term persisting vaccine-specific CD8 T cells originate from early rapid dividers, subsequently divide less than once a year, and maintain a distinct transcriptional profile [144].

Natural immune variation

There are insights to be gained from understanding human immune variation and so-called ‘experiments of nature’. Large-scale efforts have been undertaken in recent years to quantify genetic and environmental (e.g., pathogen exposure, immunization, chronic infection, microbiome, or maternal health) factors that contribute to observed immune variation among healthy people. The relative contributions seem to vary by cell type and human populations studied, as innate immune responses have been identified as more genetically controlled compared to adaptive responses [145,146,147]. Understanding immune variation has been a particularly rich area for HIV research too, with progress made in understanding immunological features of resistance against infection despite repeated exposure to the virus, long-term viral control, and non-progression to AIDS even in the absence of anti-retroviral drugs [148, 149].

Primary immunodeficiency patients who present with a constellation of susceptibility to infectious diseases and/or autoimmunity are also a window into the more mechanistic aspects of human immunity. In one recent clinical case, CD70 deficiency was shown to have a detrimental effect on T cell responses to EBV-infected B cells [150]. Izawa et al. [150] showed that disruption of the CD27/CD70 costimulation pathway resulted in defective T cell cytolytic function and proliferation against EBV-infected B cells through a TCR-mediated process. Reconstitution of CD70 expression restored normal functional activity. Individuals with these rare inborn mutations and their subsequent treatment have revealed a great deal about cell signaling in human immune cells and host–pathogen interactions in exquisite detail.

The flow cytometer and CDs

The combination of fluorochrome-conjugated monoclonal antibodies with the timely invention of the flow cytometer/Fluorescence-activated cell sorter (FACS) by Len Herzenberg and his colleagues revolutionized immunology [41]. For the first time, large numbers of individual cells within a population of mononuclear cells from blood, or spleen and lymph nodes could be identified and separated. Stuart Schlossman together with Ellis Reinherz and Jerry Ritz and their colleagues reported that monoclonal antibodies could be used to identify functional subsets of lymphocytes and leukemia cells [42–44], and they promoted cooperative efforts that led many investigators over the course of the next decade to identify hundreds of monoclonal antibodies that react with cell surface molecules or Clusters of Determinants (CD) that detect and discriminate separate lymphocyte subsets and functions. It is now routine to send blood to the clinical lab to determine the differential count of CD3+ T cells (both CD4+ and CD8+ subsets) [44, 45], B cells (CD20+) [46], NK cells (CD16/56+) [47] and monocytes (CD14+) [48], which is performed using the flow cytometer and some of the first lymphocyte subset-specific monoclonal antibodies reported.

How antibodies work to destroy invaders

An antibody does not itself destroy microorganisms. Instead, the antibody that has been made in response to a


Antibody response — The specific immune response that utilizes B cells to neutralize bacteria and free viruses.

Antigen-presenting cell (APC) — A macrophage that has ingested a foreign cell and displays the antigen on its surface.

B lymphocyte — Immune system white blood cell that produces antibodies.

Cell — mediated response-The specific immune response that utilizes T cells to neutralize cells that have been infected with viruses and certain bacteria.

Complement system — A series of 20 proteins that complement the immune system complement proteins destroy virus-infected cells and enhance the phagocytic activity of macrophages.

Cytotoxic T cell — A T lymphocyte that destroys virus-infected cells in the cell-mediated immune response.

Helper T lymphocyte — The ’ lynch pin ’ of specific immune responses helper T cells bind to APCs (antigen-presenting cells), activating both the antibody and cell-mediated immune responses.

Inflammatory response — A non-specific immune response that causes the release of histamine into an area of injury also prompts blood flow and immune cell activity at injured sites.

Lymphocyte — White blood cell.

Macrophage — An immune cell that engulfs foreign cells.

Major histocompatibility complex (MHC) — The proteins that protrude from the surface of a cell that identify the cell as self.

Memory cell — The T and B cells that remain behind after a primary immune response these cells swiftly respond to subsequent invasions by the same microorganism.

Natural killer cell — An immune cell that kills infected tissue cells by punching a hole in the cell membrane.

Neutrophil — An immune cell that releases a bacteria-killing chemical neutrophils are prominent in the inflammatory response.

Nonspecific defenses — Defenses such as barriers and the inflammatory response that generally target all foreign cells.

Phagocyte — A cell that engulfs another cell.

Plasma cell — A B cell that secretes antibodies.

Primary immune response — The immune response that is elicited when the body first encounters a specific antigen.

Secondary immune response — The immune response that is elicited when the body encounters a specific antigen a second time due to the presence of memory cells, this response is usually much swifter than the primary immune response.

Specific defenses — Immune responses that target specific antigens includes the antibody and cell-mediated responses.

Suppressor T cell — T lymphocytes that deactivate T and B cells.

T cells — Immune-system white blood cells that enable antibody production, suppress antibody production, or kill other cells.

Vaccination — Inducing the body to make memory cells by artificially introducing antigens into the body.

specific microorganism binds to the specific antigen on its surface. With the antibody molecule bound to its antigen, the microorganism is targeted by destructive immune cells like macrophages and NK cells. Antibody-tagged microorganisms can also be destroyed by the complement system.


Sprent J, Surh CD: T cell memory. Annu Rev Immunol. 2002, 20: 551-579. 10.1146/annurev.immunol.20.100101.151926.

Ahmed R, Gray D: Immunological memory and protective immunity: understanding their relation. Science. 1996, 272: 54-60. 10.1126/science.272.5258.54.

Harty JT, Tvinnereim AR, White DW: CD8+ T cell effector mechanisms in resistance to infection. Annu Rev Immunol. 2000, 18: 275-308. 10.1146/annurev.immunol.18.1.275.

Hoft DF: Tuberculosis vaccine development: goals, immunological design, and evaluation. Lancet. 2008, 372: 164-175. 10.1016/S0140-6736(08)61036-3.

Plotkin SA: Vaccines: correlates of vaccine-induced immunity. Clin Infect Dis. 2008, 47: 401-409. 10.1086/589862.

Kawai T, Akira S: Toll-like receptor and RIG-I-like receptor signaling. Ann N Y Acad Sci. 2008, 1143: 1-20. 10.1196/annals.1443.020.

Pedra JH, Cassel SL, Sutterwala FS: Sensing pathogens and danger signals by the inflammasome. Curr Opin Immunol. 2009, 21: 10-16. 10.1016/j.coi.2009.01.006.

Takeuchi O, Akira S: MDA5/RIG-I and virus recognition. Curr Opin Immunol. 2008, 20: 17-22. 10.1016/j.coi.2008.01.002.

Geijtenbeek TB, Gringhuis SI: Signalling through C-type lectin receptors: shaping immune responses. Nat Rev Immunol. 2009, 9: 465-479. 10.1038/nri2569.

Holt LB: Developments in Diphtheria Prophylaxis. 1950, London: Wm Heinemann

Glenny AT, Pope CG, Waddington H, Wallace U: The antigenic value of toxoid precipitated by potassium alum. J Pathol Bacteriol. 1926, 29: 38-45.

Amanna IJ, Carlson NE, Slifka MK: Duration of humoral immunity to common viral and vaccine antigens. N Engl J Med. 2007, 357: 1903-1915. 10.1056/NEJMoa066092.

Simonsen O, Kjeldsen K, Heron I: Immunity against tetanus and effect of revaccination 25-30 years after primary vaccination. Lancet. 1984, 2: 1240-1242. 10.1016/S0140-6736(84)92796-X.

Sanchez J, Holmgren J: Virulence factors, pathogenesis and vaccine protection in cholera and ETEC diarrhea. Curr Opin Immunol. 2005, 17: 388-398. 10.1016/j.coi.2005.06.007.

Ogra PlK DT, Righthand F, MacGillivray M: Immunoglobulin response in serum and secretions after immunization of live and inactivated polio vaccine and natural infection. N Engl J Med. 1968, 279: 893-899.

Ellebedy AH, Webby RJ: Influenza vaccines. Vaccine. 2009, 27 Suppl 4: D65-D68. 10.1016/j.vaccine.2009.08.038.

Berard M, Tough DF: Qualitative differences between naive and memory T cells. Immunology. 2002, 106: 127-138. 10.1046/j.1365-2567.2002.01447.x.

Vesikari T, Groth N, Karvonen A, Borkowski A, Pellegrini M: MF59-adjuvanted influenza vaccine (FLUAD) in children: safety and immunogenicity following a second year seasonal vaccination. Vaccine. 2009, 27: 6291-6295. 10.1016/j.vaccine.2009.02.004.

Nicholson KG, Thompson CI, Klap JM, Wood JM, Batham S, Newman RW, Mischler R, Zambon MC, Stephenson I: Safety and immunogenicity of whole-virus, alum-adjuvanted whole-virus, virosomal, and whole-virus intradermal influenza A/H9N2 vaccine formulations. Vaccine. 2009, 28: 171-178. 10.1016/j.vaccine.2009.09.103.

Chu DW, Hwang SJ, Lim FS, Oh HM, Thongcharoen P, Yang PC, Bock HL, Drame M, Gillard P, Hutagalung Y, Tang H, Teoh YL, Ballou RW, H5N1 Flu Study Group for Hong Kong, Singapore, Taiwan and Thailand: Immunogenicity and tolerability of an AS03(A)-adjuvanted prepandemic influenza vaccine: a phase III study in a large population of Asian adults. Vaccine. 2009, 27: 7428-7435. 10.1016/j.vaccine.2009.07.102.

Schwarz TF, Horacek T, Knuf M, Damman HG, Roman F, Drame M, Gillard P, Jilg W: Single dose vaccination with AS03-adjuvanted H5N1 vaccines in a randomized trial induces strong and broad immune responsiveness to booster vaccination in adults. Vaccine. 2009, 27: 6284-6290. 10.1016/j.vaccine.2009.01.040.

He XS, Holmes TH, Zhang C, Mahmood K, Kemble GW, Lewis DB, Dekker CL, Greenberg HB, Arvin AM: Cellular immune responses in children and adults receiving inactivated or live attenuated influenza vaccines. J Virol. 2006, 80: 11756-11766. 10.1128/JVI.01460-06.

Kreijtz JH, de Mutsert G, van Baalen CA, Fouchier RA, Osterhaus AD, Rimmelzwaan GF: Cross-recognition of avian H5N1 influenza virus by human cytotoxic T-lymphocyte populations directed to human influenza A virus. J Virol. 2008, 82: 5161-5166. 10.1128/JVI.02694-07.

Lee LY, Ha do LA, Simmons C, de Jong MD, Chau NV, Schumacher R, Peng YC, McMichael AJ, Farrar JJ, Smith GL, Townsend AR, Askonas BA, Rowland-Jones S, Dong T: Memory T cells established by seasonal human influenza A infection cross-react with avian influenza A (H5N1) in healthy individuals. J Clin Invest. 2008, 118: 3478-3490.

Wang M, Lamberth K, Harndahl M, Roder G, Stryhn A, Larsen MV, Nielsen M, Lundegaard C, Tang ST, Dziegiel MH, Rosenkvist J, Pedersen AE, Buus S, Claesson MH, Lund O: CTL epitopes for influenza A including the H5N1 bird flu genome-, pathogen-, and HLA-wide screening. Vaccine. 2007, 25: 2823-2831. 10.1016/j.vaccine.2006.12.038.

Boon AC, de Mutsert G, van Baarle D, Smith DJ, Lapedes AS, Fouchier RA, Sintnicolaas K, Osterhaus AD, Rimmelzwaan GF: Recognition of homo- and heterosubtypic variants of influenza A viruses by human CD8+ T lymphocytes. J Immunol. 2004, 172: 2453-2460.

Braciale TJ: Immunologic recognition of influenza virus-infected cells. II. Expression of influenza A matrix protein on the infected cell surface and its role in recognition by cross-reactive cytotoxic T cells. J Exp Med. 1977, 146: 673-689. 10.1084/jem.146.3.673.

Jameson J, Cruz J, Terajima M, Ennis FA: Human CD8+ and CD4+ T lymphocyte memory to influenza A viruses of swine and avian species. J Immunol. 1999, 162: 7578-7583.

Kees U, Krammer PH: Most influenza A virus-specific memory cytotoxic T lymphocytes react with antigenic epitopes associated with internal virus determinants. J Exp Med. 1984, 159: 365-377. 10.1084/jem.159.2.365.

Yewdell JW, Bennink JR, Smith GL, Moss B: Influenza A virus nucleoprotein is a major target antigen for cross-reactive anti-influenza A virus cytotoxic T lymphocytes. Proc Natl Acad Sci USA. 1985, 82: 1785-1789. 10.1073/pnas.82.6.1785.

Kool M, Soullie T, van Nimwegen M, Willart MA, Muskens F, Jung S, Hoogsteden HC, Hammad H, Lambrecht BN: Alum adjuvant boosts adaptive immunity by inducing uric acid and activating inflammatory dendritic cells. J Exp Med. 2008, 205: 869-882. 10.1084/jem.20071087.

Dupuis M, Murphy TJ, Higgins D, Ugozzoli M, van Nest G, Ott G, McDonald DM: Dendritic cells internalize vaccine adjuvant after intramuscular injection. Cell Immunol. 1998, 186: 18-27. 10.1006/cimm.1998.1283.

Mosca F, Tritto E, Muzzi A, Monaci E, Bagnoli F, Iavarone C, O'Hagan D, Rappuoli R, De Gregorio E: Molecular and cellular signatures of human vaccine adjuvants. Proc Natl Acad Sci USA. 2008, 105: 10501-10506. 10.1073/pnas.0804699105.

Seubert A, Monaci E, Pizza M, O'Hagan DT, Wack A: The adjuvants aluminum hydroxide and MF59 induce monocyte and granulocyte chemoattractants and enhance monocyte differentiation toward dendritic cells. J Immunol. 2008, 180: 5402-5412.

Vyas JM, Veen Van der AG, Ploegh HL: The known unknowns of antigen processing and presentation. Nat Rev Immunol. 2008, 8: 607-618. 10.1038/nri2368.

Trinchieri G: Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol. 2003, 3: 133-146. 10.1038/nri1001.

Jakob T, Walker PS, Krieg AM, Udey MC, Vogel JC: Activation of cutaneous dendritic cells by CpG-containing oligodeoxynucleotides: a role for dendritic cells in the augmentation of TH 1 responses by immunostimulatory DNA. J Immunol. 1998, 161: 3042-3049.

Martin M, Michalek SM, Katz J: Role of innate immune factors in the adjuvant activity of monophosphoryl lipid A. Infect Immun. 2003, 71: 2498-2507. 10.1128/IAI.71.5.2498-2507.2003.

Robson NC, Beacock-Sharp H, Donachie AM, Mowat AM: The role of antigen-presenting cells and interleukin-12 in the priming of antigen-specific CD4+ T cells by immune stimulating complexes. Immunology. 2003, 110: 95-104. 10.1046/j.1365-2567.2003.01705.x.

De Gregorio E, D'Oro U, Wack A: Immunology of TLR-independent vaccine adjuvants. Curr Opin Immunol. 2009, 21: 339-345. 10.1016/j.coi.2009.05.003.

Schnorrer P, Behrens GM, Wilson NS, Pooley JL, Smith CM, El-Sukkari D, Davey G, Kupresanin F, Li M, Maraskovsky E, Belz GT, Carbone FR, Shortman K, Heath WR, Villadangos JA: The dominant role of CD8+ dendritic cells in cross-presentation is not dictated by antigen capture. Proc Natl Acad Sci USA. 2006, 103: 10729-10734. 10.1073/pnas.0601956103.

Burgdorf S, Scholz C, Kautz A, Tampe R, Kurts C: Spatial and mechanistic separation of cross-presentation and endogenous antigen presentation. Nat Immunol. 2008, 9: 558-566. 10.1038/ni.1601.

Schnurr M, Orban M, Robson NC, Shin A, Braley H, Airey D, Cebon J, Maraskovsky E, Endres S: ISCOMATRIX adjuvant induces efficient cross-presentation of tumor antigen by dendritic cells via rapid cytosolic antigen delivery and processing via tripeptidyl peptidase II. J Immunol. 2009, 182: 1253-1259. 10.4049/jimmunol.0990022.

Li H, Willingham SB, Ting JP, Re F: Cutting edge: inflammasome activation by alum and alum's adjuvant effect are mediated by NLRP3. J Immunol. 2008, 181: 17-21.

Eisenbarth SC, Colegio OR, O'Connor W, Sutterwala FS, Flavell RA: Crucial role for the Nalp3 inflammasome in the immunostimulatory properties of aluminium adjuvants. Nature. 2008, 453: 1122-1126. 10.1038/nature06939.

McKee AS, Munks MW, MacLeod MK, Fleenor CJ, Van Rooijen N, Kappler JW, Marrack P: Alum induces innate immune responses through macrophage and mast cell sensors, but these sensors are not required for alum to act as an adjuvant for specific immunity. J Immunol. 2009, 183: 4403-4414. 10.4049/jimmunol.0900164.

Kool M, Petrilli V, De Smedt T, Rolaz A, Hammad H, van Nimwegen M, Bergen IM, Castillo R, Lambrecht BN, Tschopp J: Cutting edge: alum adjuvant stimulates inflammatory dendritic cells through activation of the NALP3 inflammasome. J Immunol. 2008, 181: 3755-3759.

Franchi L, Nunez G: The Nlrp3 inflammasome is critical for aluminium hydroxide-mediated IL-1beta secretion but dispensable for adjuvant activity. Eur J Immunol. 2008, 38: 2085-2089. 10.1002/eji.200838549.

Nolte MA, Leibundgut-Landmann S, Joffre O, Reis e Sousa C: Dendritic cell quiescence during systemic inflammation driven by LPS stimulation of radioresistant cells in vivo. J Exp Med. 2007, 204: 1487-1501. 10.1084/jem.20070325.

Joshi NS, Cui W, Chandele A, Lee HK, Urso DR, Hagman J, Gapin L, Kaech SM: Inflammation directs memory precursor and short-lived effector CD8(+) T cell fates via the graded expression of T-bet transcription factor. Immunity. 2007, 27: 281-295. 10.1016/j.immuni.2007.07.010.

Kondrack RM, Harbertson J, Tan JT, McBreen ME, Surh CD, Bradley LM: Interleukin 7 regulates the survival and generation of memory CD4 cells. J Exp Med. 2003, 198: 1797-1806. 10.1084/jem.20030735.

Li J, Huston G, Swain SL: IL-7 promotes the transition of CD4 effectors to persistent memory cells. J Exp Med. 2003, 198: 1807-1815. 10.1084/jem.20030725.

Wherry EJ, Teichgraber V, Becker TC, Masopust D, Kaech SM, Antia R, von Andrian UH, Ahmed R: Lineage relationship and protective immunity of memory CD8+ T cell subsets. Nat Immunol. 2003, 4: 225-234. 10.1038/ni889.

Zinkernagel RM, Hengartner H: Protective 'immunity' by pre-existent neutralizing antibody titers and preactivated T cells but not by so-called 'immunological memory'. Immunol Rev. 2006, 211: 310-319. 10.1111/j.0105-2896.2006.00402.x.

Aagaard C, Hoang TT, Izzo A, Billeskov R, Troudt J, Arnett K, Keyser A, Elvang T, Andersen P, Dietrich J: Protection and polyfunctional T cells induced by Ag85B-TB10.4/IC31 against Mycobacterium tuberculosis is highly dependent on the antigen dose. PLoS One. 2009, 4: e5930-10.1371/journal.pone.0005930.

Appay V, Douek DC, Price DA: CD8+ T cell efficacy in vaccination and disease. Nat Med. 2008, 14: 623-628. 10.1038/nm.f.1774.

Marrack P, Kappler J, Mitchell T: Type I interferons keep activated T cells alive. J Exp Med. 1999, 189: 521-530. 10.1084/jem.189.3.521.

Badovinac VP, Harty JT: Manipulating the rate of memory CD8+ T cell generation after acute infection. J Immunol. 2007, 179: 53-63.

Badovinac VP, Porter BB, Harty JT: CD8+ T cell contraction is controlled by early inflammation. Nat Immunol. 2004, 5: 809-817. 10.1038/ni1098.

MacLennan IC, Gulbranson-Judge A, Toellner KM, Casamayor-Palleja M, Chan E, Sze DM, Luther SA, Orbea HA: The changing preference of T and B cells for partners as T-dependent antibody responses develop. Immunol Rev. 1997, 156: 53-66. 10.1111/j.1600-065X.1997.tb00958.x.

Khanolkar A, Badovinac VP, Harty JT: CD8+ T cell memory development: CD4+ T cell help is appreciated. Immunol Res. 2007, 39: 94-104. 10.1007/s12026-007-0081-4.

Locksley RM: Nine lives: plasticity among T helper cell subsets. J Exp Med. 2009, 206: 1643-1646. 10.1084/jem.20091442.

Zhou L, Chong MM, Littman DR: Plasticity of CD4+ T cell lineage differentiation. Immunity. 2009, 30: 646-655. 10.1016/j.immuni.2009.05.001.

MacLeod MK, Clambey ET, Kappler JW, Marrack P: CD4 memory T cells: what are they and what can they do?. Semin Immunol. 2009, 21: 53-61. 10.1016/j.smim.2009.02.006.

Rodrigues LC, Diwan VK, Wheeler JG: Protective effect of BCG against tuberculous meningitis and miliary tuberculosis: a meta-analysis. Int J Epidemiol. 1993, 22: 1154-1158. 10.1093/ije/22.6.1154.

Trunz BB, Fine P, Dye C: Effect of BCG vaccination on childhood tuberculous meningitis and miliary tuberculosis worldwide: a meta-analysis and assessment of cost-effectiveness. Lancet. 2006, 367: 1173-1180. 10.1016/S0140-6736(06)68507-3.

Xu R, Johnson AJ, Liggitt D, Bevan MJ: Cellular and humoral immunity against vaccinia virus infection of mice. J Immunol. 2004, 172: 6265-6271.

Yammani RD, Pejawar-Gaddy S, Gurley TC, Weimer ET, Hiltbold EM, Alexander-Miller MA: Regulation of maturation and activating potential in CD8+ versus CD8- dendritic cells following in vivo infection with vaccinia virus. Virology. 2008, 378: 142-150. 10.1016/j.virol.2008.05.031.

Goonetilleke NP, McShane H, Hannan CM, Anderson RJ, Brookes RH, Hill AV: Enhanced immunogenicity and protective efficacy against Mycobacterium tuberculosis of bacille Calmette-Guerin vaccine using mucosal administration and boosting with a recombinant modified vaccinia virus Ankara. J Immunol. 2003, 171: 1602-1609.

Williams A, Goonetilleke NP, McShane H, Clark SO, Hatch G, Gilbert SC, Hill AV: Boosting with poxviruses enhances Mycobacterium bovis BCG efficacy against tuberculosis in guinea pigs. Infect Immun. 2005, 73: 3814-3816. 10.1128/IAI.73.6.3814-3816.2005.

Beveridge NE, Price DA, Casazza JP, Pathan AA, Sander CR, Asher TE, Ambrozak DR, Precopio ML, Scheinberg P, Alder NC, Roederer M, Koup RA, Douek DC, Hill AV, McShane H: Immunisation with BCG and recombinant MVA85A induces long-lasting, polyfunctional Mycobacterium tuberculosis-specific CD4+ memory T lymphocyte populations. Eur J Immunol. 2007, 37: 3089-3100. 10.1002/eji.200737504.

Brookes RH, Hill PC, Owiafe PK, Ibanga HB, Jeffries DJ, Donkor SA, Fletcher HA, Hammond AS, Lienhardt C, Adegbola RA, Hill AV: Safety and immunogenicity of the candidate tuberculosis vaccine MVA85A in West Africa. PLoS One. 2008, 3: e2921-10.1371/journal.pone.0002921.

Hawkridge T, Scriba TJ, Gelderbloem S, Smit E, Tameris M, Moyo S, Lang T, Veldsman A, Hatherill M, Merwe L, Fletcher HA, Mahomed H, Hill AV, Hanekom WA, Hussey GD, McShane H: Safety and immunogenicity of a new tuberculosis vaccine, MVA85A, in healthy adults in South Africa. J Infect Dis. 2008, 198: 544-552. 10.1086/590185.

Darrah PA, Patel DT, De Luca PM, Lindsay RW, Davey DF, Flynn BJ, Hoff ST, Andersen P, Reed SG, Morris SL, Roederer M, Seder RA: Multifunctional TH1 cells define a correlate of vaccine-mediated protection against Leishmania major. Nat Med. 2007, 13: 843-850. 10.1038/nm1592.

Kagi D, Ledermann B, Burki K, Seiler P, Odermatt B, Olsen KJ, Podack ER, Zinkernagel RM, Hengartner H: Cytotoxicity mediated by T cells and natural killer cells is greatly impaired in perforin-deficient mice. Nature. 1994, 369: 31-37. 10.1038/369031a0.

Kagi D, Seiler P, Pavlovic J, Ledermann B, Burki K, Zinkernagel RM, Hengartner H: The roles of perforin- and Fas-dependent cytotoxicity in protection against cytopathic and noncytopathic viruses. Eur J Immunol. 1995, 25: 3256-3262. 10.1002/eji.1830251209.

Topham DJ, Tripp RA, Doherty PC: CD8+ T cells clear influenza virus by perforin or Fas-dependent processes. J Immunol. 1997, 159: 5197-5200.

Walsh CM, Matloubian M, Liu CC, Ueda R, Kurahara CG, Christensen JL, Huang MT, Young JD, Ahmed R, Clark WR: Immune function in mice lacking the perforin gene. Proc Natl Acad Sci USA. 1994, 91: 10854-10858. 10.1073/pnas.91.23.10854.

McMichael AJ, Gotch FM, Noble GR, Beare PA: Cytotoxic T-cell immunity to influenza. N Engl J Med. 1983, 309: 13-17.

Malik A, Egan JE, Houghten RA, Sadoff JC, Hoffman SL: Human cytotoxic T lymphocytes against the Plasmodium falciparum circumsporozoite protein. Proc Natl Acad Sci USA. 1991, 88: 3300-3304. 10.1073/pnas.88.8.3300.

McElhaney JE, Ewen C, Zhou X, Kane KP, Xie D, Hager WD, Barry MB, Kleppinger A, Wang Y, Bleackley RC, Granzyme B: Correlates with protection and enhanced CTL response to influenza vaccination in older adults. Vaccine. 2009, 27: 2418-2425. 10.1016/j.vaccine.2009.01.136.

McElhaney JE, Xie D, Hager WD, Barry MB, Wang Y, Kleppinger A, Ewen C, Kane KP, Bleackley RC: T cell responses are better correlates of vaccine protection in the elderly. J Immunol. 2006, 176: 6333-6339.

Jackson DC, Lau YF, Le T, Suhrbier A, Deliyannis G, Cheers C, Smith C, Zeng W, Brown LE: A totally synthetic vaccine of generic structure that targets Toll-like receptor 2 on dendritic cells and promotes antibody or cytotoxic T cell responses. Proc Natl Acad Sci USA. 2004, 101: 15440-15445. 10.1073/pnas.0406740101.

Smith RE, Donachie AM, Grdic D, Lycke N, Mowat AM: Immune-stimulating complexes induce an IL-12-dependent cascade of innate immune responses. J Immunol. 1999, 162: 5536-5546.

Sambhara S, Kurichh A, Miranda R, Tumpey T, Rowe T, Renshaw M, Arpino R, Tamane A, Kandil A, James O, Underdown B, Klein M, Katz J, Burt D: Heterosubtypic immunity against human influenza A viruses, including recently emerged avian H5 and H9 viruses, induced by FLU-ISCOM vaccine in mice requires both cytotoxic T-lymphocyte and macrophage function. Cell Immunol. 2001, 211: 143-153. 10.1006/cimm.2001.1835.

Guidotti LG, Ishikawa T, Hobbs MV, Matzke B, Schreiber R, Chisari FV: Intracellular inactivation of the hepatitis B virus by cytotoxic T lymphocytes. Immunity. 1996, 4: 25-36. 10.1016/S1074-7613(00)80295-2.

Sun P, Schwenk R, White K, Stoute JA, Cohen J, Ballou WR, Voss G, Kester KE, Heppner DG, Krzych U: Protective immunity induced with malaria vaccine, RTS, S, is linked to Plasmodium falciparum circumsporozoite protein-specific CD4+ and CD8+ T cells producing IFN-gamma. J Immunol. 2003, 171: 6961-6967.

Richie T: High road, low road? Choices and challenges on the pathway to a malaria vaccine. Parasitology. 2006, 133 Suppl: S113-S144. 10.1017/S0031182006001843.

Stoute JA, Slaoui M, Heppner DG, Momin P, Kester KE, Desmons P, Wellde BT, Garcon N, Krzych U, Marchand M: A preliminary evaluation of a recombinant circumsporozoite protein vaccine against Plasmodium falciparum malaria. RTS, S Malaria Vaccine Evaluation Group. N Engl J Med. 1997, 336: 86-91. 10.1056/NEJM199701093360202.

Schofield L, Villaquiran J, Ferreira A, Schellekens H, Nussenzweig R, Nussenzweig V: Gamma interferon, CD8+ T cells and antibodies required for immunity to malaria sporozoites. Nature. 1987, 330: 664-666. 10.1038/330664a0.

Miller JD, Most van der RG, Akondy RS, Glidewell JT, Albott S, Masopust D, Murali-Krishna K, Mahar PL, Edupuganti S, Lalor S, Germon S, Del Rio C, Mulligan MJ, Staprans SI, Altman JD, Feinberg MB, Ahmed R: Human effector and memory CD8+ T cell responses to smallpox and yellow fever vaccines. Immunity. 2008, 28: 710-722. 10.1016/j.immuni.2008.02.020.

Querec TD, Akondy RS, Lee EK, Cao W, Nakaya HI, Teuwen D, Pirani A, Gernert K, Deng J, Marzolf B, Kennedy K, Wu H, Bennouna S, Oluoch H, Miller J, Vencio RZ, Mulligan M, Aderem A, Ahmed R, Pulendran B: Systems biology approach predicts immunogenicity of the yellow fever vaccine in humans. Nat Immunol. 2009, 10: 116-125. 10.1038/ni.1688.

Theiler M, Smith HH: The use of Yellow Fever virus modified by in vitro cultivation for human immunization. J Exp Med. 1937, 65: 787-800. 10.1084/jem.65.6.787.

Clements CJ, Griffiths E: The global impact of vaccines containing aluminium adjuvants. Vaccine. 2002, 20 Suppl 3: S24-S33. 10.1016/S0264-410X(02)00168-8.

Ott G, Barchfeld GL, Van Nest G: Enhancement of humoral response against human influenza vaccine with the simple submicron oil/water emulsion adjuvant MF59. Vaccine. 1995, 13: 1557-1562. 10.1016/0264-410X(95)00089-J.

Durando P, Fenoglio D, Boschini A, Ansaldi F, Icardi G, Sticchi L, Renzoni A, Fabbri P, Ferrera A, Parodi A, Bruzzone B, Gabutti G, Podda A, Del Giudice G, Fragapane E, Indiveri F, Crovari P, Gasparini R: Safety and immunogenicity of two influenza virus subunit vaccines, with or without MF59 adjuvant, administered to human immunodeficiency virus type 1-seropositive and - seronegative adults. Clin Vaccine Immunol. 2008, 15: 253-259. 10.1128/CVI.00316-07.

Coler RN, Carter D, Friede M, Reed SG: Adjuvants for malaria vaccines. Parasite Immunol. 2009, 31: 520-528. 10.1111/j.1365-3024.2009.01142.x.

Heineman TC, Clements-Mann ML, Poland GA, Jacobson RM, Izu AE, Sakamoto D, Eiden J, Van Nest GA, Hsu HH: A randomized, controlled study in adults of the immunogenicity of a novel hepatitis B vaccine containing MF59 adjuvant. Vaccine. 1999, 17: 2769-2778. 10.1016/S0264-410X(99)00088-2.

McFarland EJ, Borkowsky W, Fenton T, Wara D, McNamara J, Samson P, Kang M, Mofenson L, Cunningham C, Duliege AM, Sinangil F, Spector SA, Jimenez E, Bryson Y, Burchett S, Frenkel LM, Yogev R, Gigliotti F, Luzuriaga K, Livingston RA, AIDS Clinical Trials Group 230 Collaborators: Human immunodeficiency virus type 1 (HIV-1) gp120-specific antibodies in neonates receiving an HIV-1 recombinant gp120 vaccine. J Infect Dis. 2001, 184: 1331-1335. 10.1086/323994.

Vélez ID, Gilchrist K, Martínez S, Ramírez-Pineda JR, Ashman JA, Alves FP, Coler RN, Bogatzki LY, Kahn SJ, Beckmann AM, Cowgill KD, Reed SG, Piazza FM: Safety and immunogenicity of a defined vaccine for the prevention of cutaneous leishmaniasis. Vaccine. 2009, 28: 329-337. 10.1016/j.vaccine.2009.10.045.

Sun HX, Xie Y, Ye YP: ISCOMs and ISCOMATRIX. Vaccine. 2009, 27: 4388-4401. 10.1016/j.vaccine.2009.05.032.

Ennis FA, Cruz J, Jameson J, Klein M, Burt D, Thipphawong J: Augmentation of human influenza A virus-specific cytotoxic T lymphocyte memory by influenza vaccine and adjuvanted carriers (ISCOMS). Virology. 1999, 259: 256-261. 10.1006/viro.1999.9765.

Lell B, Agnandji S, von Glasenapp I, Haertle S, Oyakhiromen S, Issifou S, Vekemans J, Leach A, Lievens M, Dubois MC, Demoitie MA, Carter T, Villafana T, Ballou WR, Cohen J, Kremsner PG: A randomized trial assessing the safety and immunogenicity of AS01 and AS02 adjuvanted RTS, S malaria vaccine candidates in children in Gabon. PLoS One. 2009, 4: e7611-10.1371/journal.pone.0007611.

Boland G, Beran J, Lievens M, Sasadeusz J, Dentico P, Nothdurft H, Zuckerman JN, Genton B, Steffen R, Loutan L, Van Hattum J, Stoffel M: Safety and immunogenicity profile of an experimental hepatitis B vaccine adjuvanted with AS04. Vaccine. 2004, 23: 316-320. 10.1016/j.vaccine.2004.06.006.

Giannini SL, Hanon E, Moris P, Van Mechelen M, Morel S, Dessy F, Fourneau MA, Colau B, Suzich J, Losonksy G, Martin MT, Dubin G, Wettendorff MA: Enhanced humoral and memory B cellular immunity using HPV16/18 L1 VLP vaccine formulated with the MPL/aluminium salt combination (AS04) compared to aluminium salt only. Vaccine. 2006, 24: 5937-5949. 10.1016/j.vaccine.2006.06.005.

Morefield GL, Sokolovska A, Jiang D, HogenEsch H, Robinson JP, Hem SL: Role of aluminum-containing adjuvants in antigen internalization by dendritic cells in vitro. Vaccine. 2005, 23: 1588-1595. 10.1016/j.vaccine.2004.07.050.

Rimaniol AC, Gras G, Verdier F, Capel F, Grigoriev VB, Porcheray F, Sauzeat E, Fournier JG, Clayette P, Siegrist CA, Dormont D: Aluminum hydroxide adjuvant induces macrophage differentiation towards a specialized antigen-presenting cell type. Vaccine. 2004, 22: 3127-3135. 10.1016/j.vaccine.2004.01.061.

Del Giudice G, Hilbert AK, Bugarini R, Minutello A, Popova O, Toneatto D, Schoendorf I, Borkowski A, Rappuoli R, Podda A: An MF59-adjuvanted inactivated influenza vaccine containing A/Panama/1999 (H3N2) induced broader serological protection against heterovariant influenza virus strain A/Fujian/2002 than a subunit and a split influenza vaccine. Vaccine. 2006, 24: 3063-3065. 10.1016/j.vaccine.2006.01.015.

Atmar RL, Keitel WA, Patel SM, Katz JM, She D, El Sahly H, Pompey J, Cate TR, Couch RB: Safety and immunogenicity of nonadjuvanted and MF59-adjuvanted influenza A/H9N2 vaccine preparations. Clin Infect Dis. 2006, 43: 1135-1142. 10.1086/508174.

Stephenson I, Bugarini R, Nicholson KG, Podda A, Wood JM, Zambon MC, Katz JM: Cross-reactivity to highly pathogenic avian influenza H5N1 viruses after vaccination with nonadjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a potential priming strategy. J Infect Dis. 2005, 191: 1210-1215. 10.1086/428948.