Of Brain and Toxoplasma

Posted by Kasra

This  wonderful comic had stayed in my drafts folder for a very long time and I had decided finally not to post since Calamaties of Nature had stopped publishing comics. But then again, a recent PLOS ONE paper reminded me of it again.  Toxoplasma gondii is one of my favorite parasites. It is one of the most common parasites of humans and in majority of cases lays dormant throughout life, making it one of the most successful parasites in Nature. After infection (eating poorly cooked infected meat or contact with feces of an infected cat), T. gondii escapes the gut and migrates to the brain. At first glance, it does not seem to do much over there, at least nothing drastic. But studies including current work by Ingram et al. have shown that T. gondii infection can permanently alter animal behavior, permanently meaning the behavioral change stays even after infection has been cleared.  This recent study is another example demonstrating this ability of the parasite:

Ingram et al. checked how the behavior of an infected or uninfected mouse differ when exposed to predators, in this case Bobcat urine. As you can see in the figure below (Part A), they set up a field, one part of which was spotted with Bobcat urine or Rabbit urine (controlling for effect of just urine versus predator urine). Infected, uninfected and infected with a attenuated parasite (which could successfully clear the infection) mice were let in the area to see which spots they spend more or less time at, translated to which areas they avoid and which areas they do not. As you can see in sections Bi and Bii, the mouse partly avoided the section with rabbit urine (hence the importance of controls!) but this avoidance was way stronger when exposed to Bobcat urine. Surprisingly this behavior almost vanishes in infected mice, regardless of the strain, in other words, regardless of presence or absence of parasites in the brain.

From Ingram et al. PLOS ONE, 2013

There is a body of research on possible effects of T. gondii infection on human brain and behavior and correlations to increased suicide and schizophrenia have been suggested, although not confirmed. Still we all keep in mind that correlation does not mean causation.

Could this alteration of animal behavior also have an evolutionary explanation/impact on Toxoplasma’s life cycle? In other words, is there more to this phenomenon than just parasite infects brain, brain acts strange?

There is a difference between cats (big and small, domestic and wild) and the rest of the animals when it comes to Toxoplasma. In other animals, as I mentioned above, Toxoplasma escapes the gut after ingestion and moves  to muscle  and brain tissue and just stays there. So basically the infection is kind of a dead-end. It cannot be transferred to the next host until the current host dies and gets eaten up by another one. However,cats are the main hosts for Toxoplasma. That is where the parasite goes through the sexual stage of its life cycle. Also, Toxoplasma cysts are constantly shed from an infected cat through its feces. Not a dead-end infection. Now the loss of predator in Toxo-infected mice makes sense! It helps the parasite get back to its main host where it can complete the cycle! This is a trait that would have been highly favored by natural selection whenever it evolved. Because it would strongly increase the chance of the parasite being passed around to the next host.

At the end, what the comic made me think of was, could those parasites also have parasites that would alter their behavior in their own benefit? Is this is a classic example of the Extended Phenotype idea introduced by Richard Dawkins?

Reference:

Ingram WM, Goodrich LM, Robey EA, & Eisen MB (2013). Mice Infected with Low-Virulence Strains of Toxoplasma gondii Lose Their Innate Aversion to Cat Urine, Even after Extensive Parasite Clearance. PloS one, 8 (9) PMID: 24058668

An infographic about Lyme disease

Posted by Kasra

There has been too much work overload in the current days, not letting me blog as often as I want to. Meanwhile, have a look at and share this well done infographic about Lyme disease. Click here for the source, more info and also other infographics.

Source: Nursing School Hub

A link between Il-1beta and helminth chronicity

Posted by Kasra

In contrary to many bacterial infections (TB put aside) helminth infections have the reputation of being chronic, not causing much short term damage and manipulating the host immune response. A Th2 response is canonically considered to be protective against helminth infection. However, there is a twist. In a recent publication in PLoS Pathogens, Zaiss et al. show that the worm alters the strength of the immune response a

Zaiss et al.  infect mice with Heligmosomoides polygyrus bakeri , which is a murine helminth parasite with phenotypes similar to those of humans. Within a few days they observe increased levels of IL-1beta in the peritoneum as well as the small intestine tissue. IL1-beta is a strong and tightly regulated pro-inflammatory cytokine. They authors suggest this cytokine to be produced by macrophages and DCs in the in the intestine. When infecting IL1-beta-/- mice the worm burden and amount of eggs shed in the feces drops drastically.

When purifying  spleen and mesenteric lymph node T cells from H. polygyrus infected IL1beta-/- mice, they see a larger population of IL-4 and IL-13 producing cells compared to wildtype mice, suggesting a stronger Th2 immune response. Following this up, they also observe an increase in IL-25 and IL-33, other Th2 cytokines. The authors show that the intestinal epithelial cells express the receptor for IL-1beta and could be producing these cytokines. Surprisingly, looking at IL-25 and IL-33 knock-out mice shows that only IL-25 reduces worm burden and not IL-33. Finally, the authors show that IL-1beta can attenuates the type 2 innate lymphoid cells (ILC2) which also contribute to the Th2 response to the infection. Together the authors suggest that induction of IL-1beta by the helminth dampens the Th2 response and results in helminth chronicity.

A schematic of proposed mechanism for immune response to H. polygyrus bakeri. From Zaiss MM et al., PLoS Pathog 9(8): e1003531. doi:10.1371/journal.ppat.1003531

As I mentioned above, IL1-beta secretion is very tightly regulated. A Pathogen-Associated-Molecular Pattern (PAMP) is required for its expression and a secondary signal through the inflammasome is required for its maturation and release. It appears that H. polygyrus provides both of the signals. Observing inflammasome activation by helminths is rather new. Doing a quick Pubmed search for  ‘inflammasome AND helminth’ I only found two papers, but I guess the list would grow. Pinning down the mechanisms by which the worms manage to activate this pathway would be very interesting, and also we would need to see if similar to H. polygyrus other worms also abuse this pathway in their benefit.

Zaiss MM, Maslowski KM, Mosconi I, Guenat N, Marsland BJ, & Harris NL (2013). IL-1β Suppresses Innate IL-25 and IL-33 Production and Maintains Helminth Chronicity. PLoS pathogens, 9 (8) PMID: 23935505

Helminths release anti-microbial peptide-like molecules that are immunomodulatory

Posted by Kasra

In this paper, the authors have studied peptides that are found secreted by helminths Schistosoma mansonai and Fasciola hepatica and closely resemble mammalian antimicrobial peptides, cathelidins to be precise.

First, the backgrounds: S. mansonai and F. hepatica are both trematodes or flukes. Their cyst form can be ingested via contaminated food or water and in the gut they hatch and can migrate to the liver. Like many other parasites, they don’t kill but cause morbidity. They are categorized as neglected tropical diseases and their infections are treatable and obviously preventable.

Anti-microbial peptides are very diverse in sequence and functions among different organisms, but there are similarities in their secondary and tertiary structures. They are produced by many multicellular organisms and their activities can range from bacteriocidal to immune modulatory. Some peptides can have both functions simultaneously,

Thivierge et al. start their paper by an interesting notion: the similarity of the innate immune response to helminths with the immune response to wounds and tissue injury. They are both anti-inflammatory and pro-Th2. Skewing the immune response from Th1 to Th2 and thus leading to less pathology as well as parasite chronicity is a recurring theme in parasite immunology. Read a full review here.

The peptides studied in this paper have similar secondary structure to alpha helical mammalian antimicrobial peptides (Cathelicidins) such as LL-37 and BMAP-28. A feature of these peptides is presence of amphi-pathic helices (hydrophobic on one side and hydrophilic on another side). This was also seen in the predicted secondary structures of peptides that were found to be secreted from S. mansonai and F. hepatica. 

Following this, the authors studied the peptides for a variety of anti-microbial and toxicity activities that are seen with mammalian peptides and found none to be present even at high doses (things such as pore formation, . However, what they did find was the peptides’ ability to modulate functions of immune cells. In this particular case they report inhibition of TNF secretion by macrophages and alteration of antibody secretion by B cells.

Similar secondary structure among mammalian and helminth peptides. (A) shows mammalian peptides with hydrophilic areas marked green and hydrophobic areas marked red. The dotted line and arrows in (B) show hydrophobic patches in the helmnith peptides. From Thivierge et al. 2013. PLoS Negl Trop Dis 7(7): e2307. doi:10.1371/journal.pntd.0002307

What the authors argue from their results is that the similar structure of these peptides to mammalian peptides and yet lack of toxicity allows them to effectively manipulate the immune response in their favor. These modulations could help in blunting of a strong Th1 response with lots of damage to the parasite as well as the host tissue and a milder response leading to parasite chronicity. Knocked-out parasites will better show the extent of importance of these peptides. Nontheless, longterm co-evolution of host and parasites has given rise to these peptides: they are nontoxic and modulatory at least in vitro.  This means plenty of potential in biotech and pharmaceutics!

Thivierge K, Cotton S, Schaefer DA, Riggs MW, To J, Lund ME, Robinson MW, Dalton JP, & Donnelly SM (2013). Cathelicidin-like Helminth Defence Molecules (HDMs): Absence of Cytotoxic, Anti-microbial and Anti-protozoan Activities Imply a Specific Adaptation to Immune Modulation. PLoS neglected tropical diseases, 7 (7) PMID: 23875042

Immunology meets epidemiology: A closer look at super spreaders

Posted by Kasra

Recent work by Gopinath et al. published in PLoS Pathogens touches a crucial issue in epidemiology of infectious diseases. We tend to look at infectious diseases as infecting everyone more or less uniformly. But many studies have shown that a great extent of heterogeneity exists in the amount of pathogen that is spread from infected individuals. The common rule is a 80-20 rule: it says that 20% of infected individuals account for 80% of disease spread. To learn more about this phenomenon and the research behind it, you can have a look at this comment published in Nature news and views. This rule appears to stand for many infectious diseases capable of causing epidemics. Therefore, studying and identifying the super spreaders –  or as this study puts them super shedders –  is of critical importance.  This study has tried to shed some light on the differences of the immune system of these individuals compared to normal shedders.

The authors use a special model of Salmonella enterica infection of mice where longterm infection was sustainable without the mice succumbing to disease. They then compared mice that very high numbers of bacteria present in their stool (two logs higher than what is normally expected of moderate shedders) with normally infected mice.

Comparing the immune response between moderate and supper shedders, the authors find that although the bacterial load in the gastrointestinal tract is highly different between the two groups of mice, the amount of bacteria in systemic organs such as the spleen is the same, so is the overall appearance of the mice. However, they find an enhanced innate immune response in the supper shedders marked with higher IL-6, a pro-inflammatory cytokine, in the serum and more neutrophils in the blood and other organs. On the other hand, they observe a reduction in the ratio of Th1 lymphocytes to regulatory T cells. The authors also found Th1 cells to be less responsive to proliferation-inducing cytokine IL-2.

Interestingly, the authors were able to induce some of the super shedder ‘characteristics’ by giving streptomycin to moderate shedder mice, probably giving room for Salmonella to expand in the altered microbiota. This further suggests that the observed immune characteristics are due to the bacterial load present in the gut. Finally, they look at the possible connection between neutrophil expansion and reduction in Th1 responsiveness.

Immunological differences between super shedder and moderate shedder Salmonella infected mice

The authors argue that to the be able to contain such a high bacterial load in the body, the organism would either have a weaker inflammatory response, or a way to dampen adaptive response to reduce the cost of high inflammation to the body. This is among the first steps in understanding the nature of the immune response in super spreaders. It would be interesting to see if this is a general theme in all host-pathogen interactions when super spreading occurs, or if it is different with every pathogen. This kind of research can importantly lead to better understanding of the immune response and hopefully also molecular markers that would help rapid identification the super spreaders and better control of disease outbreaks.

Gopinath S, Hotson A, Johns J, Nolan G, & Monack D (2013). The Systemic Immune State of Super-shedder Mice Is Characterized by a Unique Neutrophil-dependent Blunting of TH1 Responses. PLoS pathogens, 9 (6) PMID: 23754944

Galvani AP, & May RM (2005). Epidemiology: dimensions of superspreading. Nature, 438 (7066), 293-5 PMID: 16292292

A new mechanism of immune evasion by Leishmania parasites

Posted by Kasra

It is always pleasant to read papers published by friends and old colleagues.  Also it was about time I would write another post about Leishmania. This new paper published in Cell host and microbe (pubmed index) discusses a new mechanism that Leishmania parasites use to evade the host immunity.

 A quick reminder, the promastigote forms of Leishmania enter the mammalian host via bite of the sandfly. They are readily phagocytosed by macrophages. Yet, they manage to evade to propagate within the macrophage thanks to multiple mechanisms of immune modulation and evasion. 

Previous publications of this group that showed how Leishmania is able to impair maturation of the phagolysosome (for example delay acidification) by interfering with phagolysosome associated proteins. Their previous work had pointed to Leishmania‘s surface lipophosphoglycan LPG. Following their work on the phagosome, they look to see if there are other molecules that are also altered after Leishmania infection, stumbling upon VAMP3 and VAMP8, two SNAREs that are cleaved after infection. They find that this time this time this cleavage is due to another important virulence factor of Leishmania namely surface protease GP63.  This protease has been shown to have many immunomodulatory properties and cleaving many important macrophage proteins (phosphatases, transcription factors…) and now there is a new one on the list.  The authors show that VAMP8 is important for cross-presentation of antigens from MHCII to MHCI. To show this they create ovalbumin expressing expressing L. major and see how presence or absence of GP63 could affect activation of OT-II (ovalbumin-reactive T cells) after coculture with macrophages. Importance of VAMP8 in cross presentation is also shown by using VAMP8-/- cells.

SNAREs are important in vesicle transport and fusion. Therefore they can be targeted by pathogens like Leishmania to impair effective pathogen killing (Image from Wikipedia).

 

What is nice about this study is that by studying host-parasite interactions and immune modulation, it also helps us learn more about the innate immune mechanisms and communication of the innate and adaptive immune systems. 

 

Matheoud D, Moradin N, Bellemare-Pelletier A, Shio MT, Hong WJ, Olivier M, Gagnon E, Desjardins M, & Descoteaux A (2013). Leishmania Evades Host Immunity by Inhibiting Antigen Cross-Presentation through Direct Cleavage of the SNARE VAMP8. Cell host & microbe, 14 (1), 15-25 PMID: 23870310

Olivier M, Atayde VD, Isnard A, Hassani K, & Shio MT (2012). Leishmania virulence factors: focus on the metalloprotease GP63. Microbes and infection / Institut Pasteur, 14 (15), 1377-89 PMID: 22683718
 

Macrophages commit ‘defensive suicide’ after Adenovirus and Listeria infection

Posted by Kasra

Cells often kill themselves for the benefit of their lot. New forms of cell suicide are being discovered every day now.  I wrote about apoptosis, which is a rather clean form of cell suicide recently. However, necrosis which until recently seemed to be a an uncontrolled form of cell death, is now being looked at again as a form of controlled suicide. A recent publication by  Di Paolo et al in the new journal of Cell Reports sheds some light on on of these rather unusual forms of cell death. The authors call it ‘defensive suicide’.

Di Paolo et al. intravenously injected Adenovirus into the mice. They observed that the macrophages (specifically in this paper, liver macrophages) capture the virus particles. However, shortly after the macrophages died of necrosis. Interestingly, they find this phenomenon to be independent from normal mediators of cell death such as various Caspases, as well as inflammatory mediators such as MyD88, TRIF and ASC. They finally point to IRF3,  a transcription factor normally activated after certain infections. Macrophages from IRF3-/- mice did not go through necrotic death after Adenovirus infection. The authors next show the proteins upstream of IRF3 are dispensable for the necrotic death of macrophages and that IRF3 is not phosphorylated at the time of macrophage necrosis, further adding to the enigma of the mechanism. The only clue we get so far is that this mechanism is dependent on escape the of the pathogen from the phagolysosome into the cytosol. They show this by using Adenovirus and also Listeria monocytogenes  mutants that cannot escape the phagolysosome. Compared to their wildtype counterparts, the mutant intracellular pathogens do not induce necrotic death of the macrophages.

Finally, to see if this necrotic death actually has a benefit for the host, the authors deplete mice from macrophages and infect them again with Adenovirus or L. monocytogenes. They observe that without the macrophages the virus or bacterial burden is a lot higher in the liver. Thus, this mechanism could be a way of slowing down the systemic spread of infection. The macrophages might collect the pathogens that would be otherwise infecting other defenseless cells and destroy them via necrotic death. Would this mean that necrotic death better kills the intracellular pathogens compared to other forms of programmed death? Or they just go through this pathway because other pathways of programmed death are blocked by the pathogens? Considering that necrosis occurs very rapidly (within minutes), the first one seems more likely.

The possible role of IRF3 in induction of necrotic death in macrophage following intracellular infection. From Di Paolo et al. , Cell Reports, Volume 3, Issue 6, 1840-1846, 13 June 2013

This mass suicide of macrophages is a very interesting phenomenon. It also raises many questions that have not yet been addressed. The most obvious question is the signaling mechanism by which IRF3 induces this special form of necrosis. The authors did not find any dependence on the proteins that are usually known to be upstream of IRF3. So there might be a novel mechanism involved. Another question concerns the route of infection. The authors have used intravenous injection both for Adenovirus as well as L. monocytogenes infection. However, these pathogens usually enter the body from the gut or the lungs and then reach the circulation system. Would this defensive necrosis extend to the immune cells in other tissues such the lung or the gut macrophages? Would the route of infection affect the intensity/quality of macrophage necrosis? We will hopefully get the answers in the near future!

Di Paolo NC, Doronin K, Baldwin LK, Papayannopoulou T, & Shayakhmetov DM (2013). The Transcription Factor IRF3 Triggers “Defensive Suicide” Necrosis in Response to Viral and Bacterial Pathogens. Cell reports, 3 (6), 1840-6 PMID: 23770239

A new murine model of Giardia infection: linking pathogenesis to malnutrition

Posted by Kasra

Giardia is a very successful parasite.   It’s highly durable cysts enter the body via contaminated food. Once inside the small intestine, the cysts hatch and the trophozoites start swimming with their multiple flagella. They attach to the intestine’s surface and enjoy the nutrient rich environment of the small intestine.  Shortly after, they start producing cysts that leave the body via feces.  Their presence may or may not cause severe symptoms. In many cases people carrying Giardia in their body – thus shedding cysts –  might not feel anything. On the other hand, Giardia infection can lead to severe diarrhea that could stick around for weeks if untreated. Our immune system usually manages to control the infection and get rid of the parasite. But in any case, the parasite can come, live and go undetected.

Similar to many other parasitic diseases, since Giardia infection does not kill a significant number of its victims, is not rampant in industrialized countries and is more or less readily treatable, it is not funded and studied by many researchers. Unfortunately, despite unique biological features such as lack of mitochondria, presence of two nuclei and an anti-inflammatory host-pathogen interaction, Giardia remains largely understudied.  We don’t fully understand the host and pathogen factors that could lead to disease or just sub-clinical infection. Nor we know much about how Giardia gets detected by the immune system and what is the nature of the immune response that kicks the parasite out of the body.

A scanning electron micrograph of the surface of the small intestine of a gerbil infested with Giardia sp. protozoa. The intestinal epithelial surface is almost entirely obscured by the attached Giardia trophozoites. (Source Wikipedia)

One of the strong incentives for studying Giardia is its higher prevalence in children. Infectious diarrhea is still the most common of cause of child death (Cryptosporidium is another understudied parasite and causative agent of diarrhea in children, which I discussed in another post). A recent study by Bartelt et al. published in Journal of Clinical Investigation, presents a new model of Giardia infection, focusing on malnutrition and young age. They argue that many children in areas where Giardia infection is common are undernourished. This malnutrition could contribute to development of a persistent Giardia infection with severe symptoms rather than a shorter non-symptomatic infection. To study the effect of children malnutrition on Giardia infection, they set their model on 3-week old recently weaned (taken away from mother, eating solid food) mice. They show that although healthy mice are able to clear the infection, Giardia parasites manage to stay longer in the small intestine of malnourished mice and also cause more growth impairment. It can be thought that this is a vicious cycle, where infectious diarrhea causes further weakening of the individual and thus further difficulty in fighting the infection, leading to severe weight loss and persistence of the parasite. Interestingly, the load of parasite in the intestine remains unchanged when comparing healthy and malnurished mice. However, the authors describe their model by pointing to other differences such as immune response and small intestine pathology. More studies on this model can help us better understand and hopefully better treat Giardia infection in children.

Highlight in Nature reviews in Gastroenterology and Hepatology

Bartelt, L., Roche, J., Kolling, G., Bolick, D., Noronha, F., Naylor, C., Hoffman, P., Warren, C., Singer, S., & Guerrant, R. (2013). Persistent G. lamblia impairs growth in a murine malnutrition model Journal of Clinical Investigation, 123 (6), 2672-2684 DOI: 10.1172/JCI67294

Exosomes derived from grapes communicate with intestinal stem cells

Posted by Kasra

I worked on exosomes for some time, so I have written about them and other secreted vesicles every now and then. I still try to follow up the research in the field and get excited with the new findings, methods and applications. Here is exciting work by by Ju et al.  where they study the interaction between grape (yes the fruit) exosomes with mouse intestinal cells.

They purified exosome-like vesicles from grapes that they bought from grocery stores. Given to mice, the exosomes appeared to be absorbed by intestinal stem cells present in the intestinal crypts. Interestingly, the authors show that picking up these vesicles induce the Wnt/β-Catenin pathway. Generally, activation of this pathway promotes proliferation of stem cells. Ju et al. show grape exosomes induce proliferation of stem cells by putting stem cells together with the vesicles ex vivo and looking at crypt formation in organoids. Finally, to provide a health-related application for their vesicles, the authors show that grape exosomes protect mice and delay death in a murine colitis model. This can be because the vesicles induce proliferation of the stem cells and thus enhance tissue regeneration to revert the damage caused by colitis.

Formation of crypts from a single intestinal stem cell ex vivo is quickened when exposed to grape exosomes. From Ju et al. 11 June 2013;doi: 0.1038/mt.2013.64

This study contains so many buzz-words, I could think of terrible ways by which it can be mis-interpretted by media: grapes heal gut disease, nano-particles in fruit protect against gut disease, … As the authors say in the first paragraphs of the paper, this study is a proof of concept and there is still a lot more to learn. Thankfully, they show enough evidence to tickle other scientists to look at application of plant-derived exosomes and exosome-like vesicles as means for drug delivery and therapy. For one thing, plants-derived products are available in higher abundance compared to their animal-derived counterparts and can be cheaper to purify in commercial quantities.  Also, humans have been exposed to them (maybe not in such high concentrations) for millions of years, so they are no strangers to the gut cells.

Ju S, Mu J, Dokland T, Zhuang X, Wang Q, Jiang H, Xiang X, Deng ZB, Wang B, Zhang L, Roth M, Welti R, Mobley J, Jun Y, Miller D, & Zhang HG (2013). Grape Exosome-like Nanoparticles Induce Intestinal Stem Cells and Protect Mice From DSS-Induced Colitis. Molecular therapy : the journal of the American Society of Gene Therapy PMID: 23752315

Bacteriophages may protect us against pathogens

Posted by Kasra

Given the extremely large amount of bacteria in our gastrointestinal track, it is not surprising to think that the gut would be also swarming with pathogens of bacteria, that is bacteriophages as well. In their recent work published in PNAS, Barr et al. take a look at what impact these particles could have on the population of bacteria in mucosal surfaces and what could it mean for us. Their work actually turns out very interesting results.

Mucosal surfaces are the body’s points of contact  with the outside. Being highly populated with bacteria, they can be suitable points of infection as well. That is why they are heavily guarded with various immune barriers and mechanisms, both innate and adaptive. Barr et al. point to a possible mechanism of protection against infection which not innate nor adaptive. They start by comparing the amounts of bacteria and bacteriophages in different mucosal and non-mucosal surfaces in various mucus producing animals. They interestingly observe that the bacteriophage to bacteria ratio in mucosal sites is way larger than those in adjacent non-mucosal sites (from average about 40fold to average about 3fold). They verify this in both invertebrates and vertebrates and thus suggest that this could be a phenomenon in all mucus-producing metazoans.

Next, they point to a previous recent study by Minot et al. who had found immunoglobulin (Ig)-like domains in the total analyzed genome of human gut viruses (or so called human gut virome).  These domains that usually act as in recognition and binding (as an antibody would do); they show that the bacteriophages actually bind to mucus through these proteins.  Barr et al. also show that presence of bacteriophages on a mucosal surface significantly reduces Escherichia coli invasion in vitro.

Model for how presence of bacteriophage on the mucosal surface can help in protection against bacterial infection. From Barr et al. PNAS 2013 PMID: 23690590

This is an incredible system where the benefit of the bacteriophages and their hosts actually match. It is not clear at this point whether the animal body would have to do something other than producing mucus to keep the bacteriophages where they are or that it is just enjoying this protection more or less free of charge.

Barr JJ, Auro R, Furlan M, Whiteson KL, Erb ML, Pogliano J, Stotland A, Wolkowicz R, Cutting AS, Doran KS, Salamon P, Youle M, & Rohwer F (2013). Bacteriophage adhering to mucus provide a non-host-derived immunity. Proceedings of the National Academy of Sciences of the United States of America PMID: 23690590

Minot S, Grunberg S, Wu GD, Lewis JD, & Bushman FD (2012). Hypervariable loci in the human gut virome. Proceedings of the National Academy of Sciences of the United States of America, 109 (10), 3962-6 PMID: 22355105