SEARCH

SEARCH BY CITATION

Introduction: The dawn of a new era in the study of intercellular communication?

  1. Top of page
  2. Introduction: The dawn of a new era in the study of intercellular communication?
  3. “No amount of experimentation can ever prove me right; a single experiment can prove me wrong” (Einstein)
  4. “Facts are the enemy of truth” (Don Quixote) if molar concentrations of the exosomes used in experiments in vitro do not correspond to those occurring in vivo
  5. Final remarks: We are successors of many generations of physical chemists – let us keep their scientific traditions
  6. References

Conventional views of how proteins and other cellular components may traffic between eukaryotic cells have been challenged in recent years 1–7. Beyond the classical secretory/exocytic pathway, the roles of secreted or shedding microvesicles (MV) in intercellular signaling have attracted great attention. MVs are released by all cells investigated to date and are present in practically all body fluids. The diverse extracellular vesicles are collectively called membrane vesicles 8. Vesicles of 100 nm–1 mm in diameter are often referred to as microparticles or MVs, and are generated by shedding from the plasma membrane. Vesicles of 50–100 nm in size are referred to as exosomes. Exosome biogenesis differs from that of membrane vesicles and occurs in two-steps: (i) the formation of multivesicular bodies (MVBs) from endosomes, when parts of their membrane invaginate and bud into the lumen of the compartment to form intralumenal vesicles (ILVs), and (ii) MVB-plasma membrane fusion, which releases these ILVs into the extracellular milieu as exosomes 9. However, a clear and decisive experimental distinction between MVs and exosomes cannot be made 10. Here, I use the size-based classification above.

Over the last few years, exosomes have received considerable attention 8, and I use them to illustrate my concern. Exosomes contain a variable spectrum of molecules enclosed inside the vesicles and on their membrane, such as signal proteins and receptors, effector proteins, lipids, DNA, and RNA cargo. Their pattern is specific to the secreting parental cell. Exosomes are often viewed as “packages” of information that can reach various targets, representing a novel form of communication between cells. In the “classical” intercellular communication, cells secrete specific molecules serving as signals for the target cells. In contrast, exosomes might be able to deliver more signals and more information to the acceptor cells, thus allowing a more complex cellular response 11–13. An important concept is that exosomes can provide an exchange of genetic information, mainly in the form of mRNA or microRNAs 14. Theoretically, this information can be transferred over long distances.

However, the physiological relevance of exosomes has been difficult to evaluate because their origin, biogenesis, and secretion mechanisms remain enigmatic 8. The new concepts are mainly based on molecular and cellular data in vitro, but direct evidence for the role of exosomes in vivo is currently lacking. To prove this concept, one needs to have tools for specifically inhibiting or increasing exosome secretion or uptake, without affecting secretion of other membrane vesicles and general secretion of proteins or lipid mediators 1, 9, 15. So far such attempts have been unsuccessful (see reviews above). So, no way to falsify 16 the hypothesis about information transfer by exosomes has yet been found, and all the relevant information is just an accumulation of supporting data. Despite some authors' uncertainty of the real physiological role of exosomes 1, 8, 9, 17, the currently prevailing opinion is that exosomes might carry an “enhanced potential” in the information flow between cells and open a “new era” in the study of intercellular communication (see e.g. 3, 10, 11, 18–21). In addition, great practical applications have also been suggested 14.

“No amount of experimentation can ever prove me right; a single experiment can prove me wrong” (Einstein)

  1. Top of page
  2. Introduction: The dawn of a new era in the study of intercellular communication?
  3. “No amount of experimentation can ever prove me right; a single experiment can prove me wrong” (Einstein)
  4. “Facts are the enemy of truth” (Don Quixote) if molar concentrations of the exosomes used in experiments in vitro do not correspond to those occurring in vivo
  5. Final remarks: We are successors of many generations of physical chemists – let us keep their scientific traditions
  6. References

Einstein's words reflect Karl Popper's falsifiability principle 16. Although at present there is no falsifying test for the hypothesis of the physiological role of MVs, quantitative estimations can also be used to falsify conclusions, despite the scant quantitative data on the number of exosomes secreted by a cell, the concentration of exosomes in the milieu, etc. Here, I use approximate estimates to show that there are reasons to challenge at least part of the new concept, namely the transfer of information by exosomes over long distances. I advance three interconnected hypotheses:

  • (1)
    Among most appropriate candidates for transferring genetic information that may cause changes in acceptor cells are those that can be amplified, e.g. mRNAs, which can be translated repeatedly to produce a multitude of proteins and cause an effect even if only a few mRNA molecules enter into the cell.
  • (2)
    To cause a noticeable effect, the product of the mRNA translation should have a regulatory function, such as a role in transcription regulation by triggering a cascade of events needed to induce the effect. Such a product might be a transcription factor (TF).
  • (3)
    Uptake of mRNA by exosomes is not subject to strong selection. The total number of identified microvesicular mRNAs, evaluated in 22 by microarray analysis, is 11,327, although there are different estimations (e.g. 23). This number is close to the total number of genes expressed in a tissue 24; therefore, the supposition is more or less substantiated. Although some reports indicate that a kind of selection still exists (e.g. 22, 25), it cannot qualitatively change the main conclusion from my approximations.

So, let us estimate how many mRNAs of length 1,000 nucleotides (nt) could fit into an exosome with the mean diameter of 100 nm. The volume of such a particle is 4/3πR3 = 0.5 × 10−3 µm3, whereas the calculated volume of a 1,000-nt RNA 26 is 3 × 10−7 µm3. One exosome can thus tightly accommodate about 1,600 RNA molecules. Considering that the exosome contains also voluminous proteins and lipids, this number is probably considerably lower. For the sake of argument, let it be 1,000. One thousand molecules randomly selected from 11,000 mRNAs by an exosome already means that each individual exosome carries a specific mRNA cargo, different from the others.

I further illustrate this with an example of TF mRNA. It has been reported that transcriptomic content of exosomes correlates with that of the parental cells 22. The number of expressed TFs varies greatly between tissue types, but the proportion of TFs expressing genes relative to all expressed genes is stably ∼6% 27. The data of 27 make it possible to deduce that the maximum number of TF types transcribed in a tissue is about 700. In a very rough approximation, 60 of 1,000 transcripts in the exosome will be TF mRNAs (this figure is actually lower than 60 because non-TF genes have higher average expression levels than TF genes 27). A rough calculation of the number of possible combinations of 60 from 700 different mRNAs gives 1087 possible variants of exosomes in different sets of TF mRNAs. Depending on the initial assumptions, this result may be 1010 or 10240, but this is qualitatively the same: the exosome cargo is extremely variable. Each exosome is individual and unique and bears a random set of information. Exosomes can be grouped into sets that share common traits, such as surface antigens and various receptors 5, 8, but exosomes within each group differ from each other. What part of this heterogeneous exosome mixture carries the information needed to cause the desirable effect in target cells? Is this part sufficient to generate a sufficiently high signal level? The situation with exosomes is dramatically different from the classical transfer of information in physiology, where the signal is generated by identical molecules, e.g. hormones, in sufficient concentrations. It should also be remembered that exosomes are internalized by endocytosis 28 or phagocytosis 29, and then either destroyed in lysosomes, expelled back out of the cell or greatly disturb the established regulatory systems causing apoptosis. The specific exosome will join a huge pool of various extracellular exosomes and thus become “a needle in a haystack”. Therefore, one should be cautious regarding the role of exosomes in intercellular transport of genetic information. So, why do the events that we think unlikely to happen in vivo still happen in vitro?

“Facts are the enemy of truth” (Don Quixote) if molar concentrations of the exosomes used in experiments in vitro do not correspond to those occurring in vivo

  1. Top of page
  2. Introduction: The dawn of a new era in the study of intercellular communication?
  3. “No amount of experimentation can ever prove me right; a single experiment can prove me wrong” (Einstein)
  4. “Facts are the enemy of truth” (Don Quixote) if molar concentrations of the exosomes used in experiments in vitro do not correspond to those occurring in vivo
  5. Final remarks: We are successors of many generations of physical chemists – let us keep their scientific traditions
  6. References

According to Occam, the best among competing solutions to a problem is the one which offers the simplest explanation. The simplest explanation to the problem that we discuss is that the exosome concentrations used in experiments in vitro never occur in vivo. Here, I give an estimate of the intercellular exosome concentration in vivo.

Exosomes probably originate from ILVs. According to 20, “… one RBL-2H3 cell could contain 3,000 ILVs…” Therefore, the upper limit of the exosome number excreted by a cell is expected to be ∼103. About 2 × 1012 exosome particles from reticulocytes would be released per day in the blood stream 20.

Unfortunately, practically all authors conducting in vitro experiments employ weight units (µg exosome protein) or weight concentration (µg/mL) as quantitative measures of exosomes, which conceals the molar concentration of exosomes essential for physiological functions and makes it difficult to correlate the experimental results with the situation in vivo. I have calculated the number of exosomes that corresponds to 1 µg exosome protein. The size of a cell is 10–20 µm and that of exosomes 30–100 nm. Therefore, the volume and mass of an exosome should be ∼10−6 of those of the cell. The average mass of protein in the cell is 0.5 ng, and thus the average mass of protein per exosome is 0.5 × 10−6 ng, and 1 µg exosome protein corresponds to 2 × 109 exosomes (calculated using the data at http://www.ncbi.nlm.nih.gov/books/NBK21121). The estimation in 20, 21 showing that 1 mg protein corresponds to (5.96 ± 0.13) × 105 exosome vesicles is probably an accidental mistake. A recent study 30 reports an average amount of 60 µg exosomal protein in 100 mL blood. According to my estimates, this corresponds to 4.2 × 1012 exosomes in 3.5 L human blood, which is higher than 2 × 1012 exosomes produced by reticulocytes alone, but of the same order of magnitude.

Next, I use my estimates to quantitatively assess the data of a “typical” exosome paper: “MC/9 exosomes (1,000 µg) were added to HMC-1 cells (8 × 106) at three different time points (0, 3, 6 hours)…” Having converted the mass of the applied exosomal protein to the number of particles as indicated above, we can deduce that the authors used 2 × 1012 cell-specific exosomes (∼106 exosome particles per cell) – a quantity equal to the total number of exosomes in the blood plasma. It seems doubtful that such a specific exosome concentration and exosome per cell ratios can occur in the real organism. Other similar examples can be presented, and some authors use even higher concentrations of cell-specific exosomes by enriching them using the presence of specific surface markers, e.g. epithelial cell adhesion molecule (EpCAM). As a result, the events unlikely in vivo due to low exosome concentrations become feasible in vitro.

Final remarks: We are successors of many generations of physical chemists – let us keep their scientific traditions

  1. Top of page
  2. Introduction: The dawn of a new era in the study of intercellular communication?
  3. “No amount of experimentation can ever prove me right; a single experiment can prove me wrong” (Einstein)
  4. “Facts are the enemy of truth” (Don Quixote) if molar concentrations of the exosomes used in experiments in vitro do not correspond to those occurring in vivo
  5. Final remarks: We are successors of many generations of physical chemists – let us keep their scientific traditions
  6. References

Classical physiologists use the language of chemistry, physics, and biochemistry with an important unit, mole/L. So why are the researchers in the field of MV physiology inclined to use µg instead of mole? We want to build a golden castle of system biology, and at the same time disregard the basics of quantitative molecular analysis. This approach may turn the golden castle into a fairy tale mirage. It seems that the “new era” in the study of intercellular communication can be epitomized by a well-known quotation from Wolfgang Pauli: “This isn't right, this isn't even wrong”. We are witnesses to a silent renunciation of correct quantitative estimates for planning and evaluating the experimental results, which would surely make Amedeo Avogadro cry.

References

  1. Top of page
  2. Introduction: The dawn of a new era in the study of intercellular communication?
  3. “No amount of experimentation can ever prove me right; a single experiment can prove me wrong” (Einstein)
  4. “Facts are the enemy of truth” (Don Quixote) if molar concentrations of the exosomes used in experiments in vitro do not correspond to those occurring in vivo
  5. Final remarks: We are successors of many generations of physical chemists – let us keep their scientific traditions
  6. References