Researchers Use MS to Find ‘Cross-talk’ Between Autophagy and Phagocytosis

Name: Wenquin (Angel) Shui

 

Position: PhD student in chemistry, University of California, Berkeley, 2004 to present

 

Background: Research Assistant Fellowship, Lawrence Berkeley National Laboratory, 2004 to 2008; MS in analytical chemistry, Fudan University in Shanghai, 2001 to 2004.

 

A team of researchers from the University of California, Berkeley, and elsewhere recently used mass spectrometry to discover and identify 546 proteins in the membrane of a phagosome, including many that had not been previously detected, in the most comprehensive proteomic profiling of the phagosomal membrane.

 

The work builds on earlier research by Michel Desjardins at the University of Montreal and Leonard Foster at the University of British Columbia, but looks specifically at membrane-bound proteins.

 

In addition to identifying new proteins, the researchers found that the level of one particular peptide, LC3-II, considered a marker for autophagy, was increased when autophagy in macrophages was induced, but reduced when autophagy was suppressed. This, they speculated, suggests that “cross-talk” between autophagy and phagocytosis may play an important role in immune-system response.

 

In particular, the finding may shed light on how tuberculosis, a focus of study for the researchers, suppresses the immune system.

 

ProteoMonitor this week spoke with Wenquin Shui, the lead author of a paper describing the research published Oct. 29 in the online edition of the Proceedings of the National Academies of Sciences.

 

Below is an edited version of the conversation.

 

 

 

Basically, all these previous proteomic studies [looking at the] phagosome organelle have been sampling the entire content of the compartment. They didn’t specifically separate the membrane proteins from the large amount of abundant soluble proteins inside the organelle, so many of these abundant proteins such as the soluble hydrolases could possibly obscure the low-abundant membrane-bound regulatory proteins [which] have more important signaling functions.

 

Based on those methods they developed for these organelle separations, we did a further step to deplete those soluble proteins and only focused on these membrane-bound signaling molecules.

 

That really allowed us to enrich for these membrane proteins and we found some very interesting receptors in the lipid-bound molecules that are not easily seen by those previous comprehensive methods.

 

The first step [of our work] is the same [as those of earlier projects]. We all need to isolate these phagosomes first by some classical method, but then we did some additional runs using biochemical methods to specifically purify the membrane sheets from the organelle.

 

 

Because we know that many of these regulatory proteins are responsible for mediating membrane trafficking of vesicle fusions … Other functions or processes engaged by these organelles are governed by membrane-bound proteins.

 

Many of these abundant soluble ones are just hydrolases or enzymes in the vesicle but they are not the functional governor for these organelles.

 

 

Actually, the reason we used both of them is … these two instruments have different intrinsic capabilities in terms of the type of proteins they can ionize and identify, so it’s not necessarily the same pool of proteins reported by two instruments.

 

By pooling the IDs from the two instruments, we can specifically expand the number of IDs we found.

 

I didn’t do a very extensive comparison [of] the hits [from] the two instruments. … Both of them found a comparable number of proteins [between 400 and 500 each], but these IDs don’t overlap very well, so I definitely found some unique ones.

 

 

In the histogram plotted against the number of transmembrane helices [in the PNAS article], you can clearly see that those proteins with more than one transmembrane domain are represented at a higher frequency in our database than [in Foster’s], which means our method really allowed us to enrich and identify more membrane-bound proteins.

 

That’s the first thing. Then all the new interesting proteins we found and validated in Western blot analysis, most of them, like the toll-like receptors and the vesicle-associated membrane protein 4 and the light-chain 3, are all known to be membrane-bound, so there is no way to identify them easily by just sampling the entire content of the phagosome.

 

 

I guess most of them are hydrolases, some enzymes localized in the lumen of the organelle, most of which are depleted in our wash after we isolate the phagosome.

 

 

This protein is a widely known marker for autophagic machinery, [which] is basically a fundamental homeostatic process many cells [use] to turn over and clean up their own cytoplasm. Most cells need to obtain nutrients or remove some damaged organelles or toxic molecules … inside the cell, so they utilize this process to clean and degrade these components.

 

People know that LC3-I is found just in the cytoplasm but not in any membrane-bound vesicles, when this pathway is activated. This precursor will be conjugated to a lipid and then inserted into the membrane of another type of vesicle called the autophagosome in the cell.

 

So this is the first time that [anyone has] shown this biochemical evidence. We found a lipid-conjugated form of LC3, which is LC3-II, in the membrane envelope of the conventional bead-containing phagosome, which could indicate there may be some cross-talk between the two pathways.

 

One pathway is conventional phagocytosis, and the other is the autophagic pathway.

 

There are not many markers used for identifying autophagy activity and LC3 is the most widely used one. I think there are a couple of other known signaling molecules, [the] Atg protein family, [which] are essential for autophagy activation, but these are more of the upstream signaling molecules that are not membrane-bound, and I don’t think they would be present in this vesicle.

 

So far, we’ve only found LC3 that is known to be related to autophagy, but another protein also showed up in our study that has some correlation to the autophagy activation, VAMP4, but its real function is not well characterized.

 

That’s a very important and intriguing point in all proteomic studies. Currently, we are looking for specific modifications in the phagosome, which is glycosylation using some unique techniques developed in our lab. But in this study we didn’t show any solid data about the modifications.

 

Basically, we’re looking at some specific glycosylations that can happen on the phagosome membrane protein, if there is any regulation going on.

 

 

I know that the Desjardin group has already published a paper providing all these phosphorylated proteins in the macrophage cells, and … they found that many of them are known to be residing in the phagosome and get phosphorylated during phagocytosis.

 

Because they’ve done this extensive work, we don’t look at that, so we’re focusing on other modifications.

 

Our lab has this particular interest in studying the activity function of tuberculosis, so now I’m looking at the possible role of a particular lipid in mediating the autophagy and phagocytosic response. … But generally speaking, autophagy could be a generic immune defense mechanism for a cell to fight against a variety of bacteria and pathogens. It’s not only restricted to Mycobacterium tuberculosis.

 

 

Totally, yes, if they can biochemically isolate the phagosome containing any type of bacteria, they can look at the specific response of the organelle to the bacteria infection.

 

A very intriguing question we’re following … is we want to know some particular bacteria products that are produced by pathogens, say mycobacteria, that may possibly mediate their autophagy response when the cells are infected by these pathogens.

 

Previous studies have shown that these autophagous responses, which are supposed to help the host cell kill the pathogen, are possibly undermined by certain pathogens when they are infecting the organelle. So now we know this potential cross-talk between the phagocytosis process and the autophagy, so we speculate the bacteria could produce some specific molecule to suppress the autophagy response.

 

We have some possible candidates in our mind and we are looking at their specific activity by some classical cell biology methods.

 

 

I think two years ago.

 

Totally. Because this current work has only used profiling techniques, in the next step, currently, we are integrating this organelle proteomic profiling with quantitative mass spec techniques. We can isotope-label different samples under defined conditions so if we treat the cells with particular factors produced by infectious microbes that are known to modulate certain phagosome properties, we can isolate the phagosome at the defined condition to purify their membranes, and then compare the phagosome membrane protein composition in different samples, using quantitative methods.

 

By this means, we expect to get more in-depth understanding of the molecular basis on how a given microbial product could mediate the phagosome functions in the host cell.

 

This will give us a clearer idea which specific proteins are regulated by a certain stimulant.

 

 

We’re using both the SILAC metabolic labeling technique and this chemical labeling [technique], iTRAQ.

 

So now because we’re using the iTRAQ in most cases … we have to use the Q-TOF instrument because it’s a little more technically demanding to use the ion trap for iTRAQ-labeled samples.