Wednesday, May 2, 2018

Lung tumors "talk" to bone cells to promote tumor growth and spread!

And I'm back!  My apologies for the long wait between posts; I've been working on a few other projects in the interim (more on this another time) and haven't had much time to write lately.

Anyway, let's dive right into the interesting science I have for you today!  I'm jumping the gun a bit and posting this write-up of an upcoming episode of Audiommunity for two reasons, 1.) because we actually recorded the episode many months ago but had to painstakingly reconstruct the audio because of technical issues with our recording software, and therefore I have been sitting on this write-up for quite some time, and 2.) I just think this work is so darn cool! 

In the general field of cancer research, the tissues, cells and organs directly surrounding tumors are referred to as the tumor "microenvironment", and understandably is a hot area of study.  Having a good grasp on what the non-tumor "bystander" cells are doing near the tumor helps researchers develop better-targeted therapies (ie turning the immune system against the tumor, à la immunotherapy).  But what is less well understood is whether or not tumors have any systemic effects on organs and tissues more distant from the primary cancer site, and whether those distant tissues may in turn influence the tumors growth and behavior.  

To get at this question, Audiommunity's very own Camilla Engblom, along with coauthor Christine Pfirschke, and a large team directed by Mikael Pittet at Mass General Hospital in Boston, investigated a specific type of lung cancer called lung adenocarcinoma and discovered that, remarkably, the tumors "talk" to cells in the bone marrow (in mice).  The result of this cross-talk is the trafficking of cells to the lungs that promote tumor growth and spread.  This work was published in the journal Science last December as a paper entitled "Osteoblasts remotely supply lung tumors with cancer-promoting  SiglecFhigh neutrophils".  

So how did they make this important discovery?  It began with a simple observation that mice with lung tumors had increased bone density and bone formation activity.  They pinned that activity down to cells called osteoblasts, the cells in the bone marrow responsible for building new bone.  To test the connection between the lung tumors and osteoblasts, they created a mouse model in which the osteoblasts were fluorescently labeled (so they could be tracked) and were able to be depleted in the mouse upon injection of diphtheria toxin.  When they knocked down the osteoblasts in mice with lung tumors, they found that over time the tumors were reduced in size and number.  

How were these bone marrow-resident osteoblasts having an impact on tumors all the way in the lung?  The researchers hypothesized that the osteoblasts must be supplying the tumor microenvironment with tumor-promoting cells.  When they looked at the type and amount of immune cell infiltrates in the lung, they found that knocking out osteoblasts resulted in a two-fold decrease of neutrophils.  Coming at it another way, if they knocked down neutrophils using an antibody without knocking down osteoblasts, they saw a large reduction in tumor volume.

These results weren't associated with all neutrophils, but specifically a population of tumor-promoting neutrophils with high levels of a lectin called SiglecF on their surface.  These SiglecF(high) neutrophils were pretty rare in normal, healthy lung tissue, but were increased 70-fold in mice with lung tumors!  Differential gene expression comparing SiglecF(high) neutrophils to SiglecF(low) neutrophils showed the SiglecF(high) cells expressed genes associated with tumor-promoting functions, such as angiogenesis, myeloid cell differentiation and recruitment, extracellular matrix remodeling, suppression of T cell responses, and tumor cell proliferations and growth, among other things (yikes!).  Additionally, when SiglecF(high) and (low) neutrophils were isolated from mice, mixed with lung tumor cells ex vivo, and then injected into new mice, tumor growth was accelerated in the mice that were injected with the SiglecF(high) neutrophil tumor preparations.  So, clearly, SiglecF(high) neutrophils= bad news.

The final piece of the puzzle is figuring out how the tumor cells are communicating with osteoblasts in the bone marrow.  A look at the serum from mice with and without tumors identified a candidate factor called RAGE (receptor for advanced glycation end products) that was upregulated about two-fold in the tumor-bearing mice. RAGE, as well as the circulating form of this receptor called sRAGE, have previously been shown to be connected with bone activity and regulation.  In good agreement with this, when they added sRAGE to the serum of tumor-free mice, they saw increased osteoblast activity and increased bone marrow neutrophil maturation.  While this part of the study will require more work to pin down, it all hangs together very nicely so far.

To summarize, lung tumor cells "talk" to bone marrow osteoblasts (possibly through a circulating factor called sRAGE), which then send tumor-promoting neutrophils to the sites of lung tumors.  So what is the takeaway from this work?  First, that osteoblast cell activity and SiglecF(high) neutrophils may be useful biomarkers in human patients, and possibly even good targets for cancer therapies.  And second, and perhaps more importantly, that it is critical to consider the entire body when studying cancer because cancer is a systemic disease!

Nice work Camilla et al!  Now please enjoy this cartoon!

Hmm, this call seems legit...




Monday, January 8, 2018

A vaccine for the common cold? Yes, please!


Happy New Year!  This post is a few weeks overdue- my apologies!  Blame it on the cold that keeps cycling through my family this month.  Anyway, here we go:

Well folks, it’s that time of year again.  Snow is falling softly from the sky, soup is bubbling on the stove, cookies are baking in the oven, and my children are… oh god, my children are sick.  Again.  In fact, they are on their second cold since November, and for the life of me I cannot keep enough tissues on hand for their little, runny noses.  And I, too, am on my second cold, and positively drowning myself in Chamomile tea.  At night, I’m nestled in my bed (thanks to NyQuil) while visions of cough drops dance through my head. 

And I. Am. Miserable.  I cannot overstate this.  All day long my head hurts and my sinuses ache.  My ears feel like they might burst and my nose alternates between dripping like a faucet and stopping up so tight I can’t breathe.  I cough so much my stomach muscles are looking like I’ve spent the last month doing crunches (hey, silver linings, right?).  And if I am lucky, I will be out of the woods on this one in, oh, about 5-7 more days.  Woof.  And at the rate we’re going this year, I might get two weeks of relatively healthy airways before I am taken down again by the next cold virus that happens along.

This time of year I turn into a real germaphobe (oh, who am I kidding?  I am a germaphobe all year).  I wash my hands so often they get chapped.  I ramp up the disinfecting of door handles, computer keyboards, and light switches.  I see germs everywhere I look.  The library.  The grocery store.  My son’s soccer games.  My husband and children’s very existence.  I spend the months from October to April just crossing my fingers and steeling myself against illness, knowing that the name of the game is ugly survival.  Because no one likes having a cold.  At best, they are inconvenient and annoying, and at worst they can put you out of commission for several days, exacerbating chronic conditions like asthma and COPD, and landing you in the hospital with pneumonia.

 
Must wash hands... Must wash hands... Must wash hands...

But researchers are fighting back by trying to create a vaccine for the common cold.  Sounds simple, right?  I mean we have vaccines for practically everything else, so you are probably asking yourself, why don’t we have one that works against colds?  There are actually a few good reasons we don’t yet have one. 

First of all, the common cold is caused by a few different types of viruses, including but not limited to rhinoviruses (the focus of this post), coronaviruses, adenoviruses, and respiratory syncytial virus, among others.  Among the rhinoviruses alone, there are 3 species (‘A’, ‘B’ and ‘C’) encompassing about 150-170 different strains that you can become infected with!  We call these different strains ‘serotypes’. That is a lot of different viruses to vaccinate against.  

 
Many of the established vaccines that we receive protect against a single serotype of a particular virus (ex the measles vaccine), or just a small number of serotypes (like the oral poliovirus vaccine, which protects against all three serotypes of poliovirus).  But the newer pneumococcal vaccine protects against 23 serotypes of virus!  Vaccines that are able to protect against multiple strains of a virus are called polyvalent vaccines, and researchers from Emory University School of Medicine in Atlanta, Georgia, are trying to create a polyvalent vaccine that would protect us against as many strains of the rhinovirus as possible.

On the most recent episode of ‘Audiommunity’ (Episode 27- Macaque PrisonGangs) we discussed a paper from this group at Emory entitled “A polyvalent inactivated rhinovirus vaccine is broadly immunogenic in rhesus macaques”, which was recently published in the journal Nature (read the original paper here).  To briefly summarize the work, the researchers made polyvalent rhinovirus vaccines by simply combining several serotypes of rhinovirus together into a vaccine cocktail, and inactivating them (so this is a “dead” vaccine, no live virus).  They then tested whether or not the cocktails could produce an immune response in mice, and later, in rhesus macaques.



Different serotypes are mixed and inactivated with formalin
The way they tested the immune response to the vaccine was by injecting it into mice and then taking a sample of the mouse serum and looking for “neutralizing antibodies”, against each of the strains of rhinovirus in the vaccine.  Neutralizing antibodies are proteins that are made by our B cells that are designed to recognize an “intruder” (in this case, the dead viruses in the vaccine), bind to it, and “neutralize” its biological effects.  A good vaccine should induce a broad neutralizing antibody response, meaning that a lot of neutralizing antibodies should be generated that recognize a lot of different parts of the “intruder”.




Mouse is injected, serum is collected, and neutralizing antibodies measured
What they found was that as long as the input amount of viruses in the cocktail was high enough (we call this the viral titer), they could get a decent immune response to the vaccine no matter how many viruses they added to it (they tested 10-, 25-, and 50-valent vaccines in this paper).  Historically, two, 10-valent rhinovirus vaccines were tested in the seventies that generated only a poor responses in test subjects (something like only 30-40% of them).  However, these vaccines had really low input virus titers that were further diluted when made into one-milliliter vaccine doses.  The researchers in this study recreated those vaccines using higher input virus titers, and were able to improve the neutralizing antibody response significantly over that of the original low titer vaccines.

The higher the input titer, the better the neutralizing antibody levels (inset: think of titer as concentration of viruses)
They went on to test a 25-valent and a 50-valent vaccine in rhesus macaques, which are a nonhuman primate common in medical research.  Remarkably, they were able to detect high levels of neutralizing antibodies against 100% of the rhinovirus serotypes in the 25-valent vaccine, and against 98% of the serotypes in the 50-valent vaccine!  This is good news for humans, since the more serotypes we can squeeze into the vaccine, the lower our chances of getting sick become.


However, there are a few minor drawbacks to this work.  The researchers found that the neutralizing antibody response is serotype-specific; there is no cross-neutralization against other viral strains not a part of a particular vaccine.  This means that an antibody against one serotype of rhinovirus is not protective against a different serotype of rhinovirus, so any serotypes not represented in a vaccine are still capable of infecting you and making you ill.  Another issue is that there were no challenge experiments in this study.  The reason for this is that there are no animal models for the common cold (you can only recapitulate some of the pathology in mice and rhesus macaques), so there is no way of knowing from this study whether these vaccines will actually prevent you from getting sick.  But this shouldn’t be too difficult to test in a clinical trial in human subjects since cold viruses aren’t all that pathogenic. 


But perhaps the biggest drawback to this work is that the researchers only tested polyvalent vaccines against the ‘A’ species of rhinovirus.  Rhinovirus A accounts for about 83 strains of virus; but B accounts for 32, and the newly discovered ‘C’ species accounts for 55 strains, which is not a small number of viruses, all vying to infect us.  The problem is that the ‘C’ species viruses are really hard to grow in cell culture.  And as this study showed, the success of the vaccine is all about getting nice, high, input virus titers.  So in the meantime, researchers are working on figuring out the best ways to grow these strains so they can add them into the vaccine cocktails. 

Emory has optioned the vaccine technology to a startup company called MeissaVaccines, Inc.  They have just received an SBIR (small business innovation research) grant from the National Institutes of Health to develop an 83-valent human vaccine, so let’s wish them luck in their efforts to make the common cold a thing of the past!  I, for one, will happily volunteer as a human test subject!