Before and after: Scientific writing

Here’s another text from a young scientist who is aiming to improve her writing. The intent of publishing it here is to show other students, teachers, and scientists the kind of work we do in communications courses. Most of the time we work individually, but here the author, Ekaterina Perets, has very generously allowed me to print the version of the text she wrote before we began working on it, alongside my comments and the final version. That text was recently published in the MDC newsletter Insights. You can see it here.

Ekaterina is a PhD student in Enno Klussmann’s lab at the MDC. As she began writing her doctoral thesis, she decided to produce a version of her abstract for non-specialists. It’s a useful exercise that challenges you to put your work into a broader perspective and look at it through other eyes.

Ekaterina wrote a solid first version of the text that we worked on for several days to produce the final draft that was published. Her goal was to tell her story clearly and accurately, in a way that would make sense to non-scientists and other researchers as well.  I have painstakingly gone through the drafts again to elaborate on some of the issues that arose and her solutions. These are things that pop up all the time in my courses on science writing. They are usually more difficult to see – and solve – in your own writing, which is why Ekaterina’s willingness to share is so valuable.

Considered in isolation, most of the problems aren’t too significant. In combination, though, they make more work for a reader who is already challenged by all the new ideas. The changes build a story that is easier to read and presents information at a pace that can be digested by a careful reader.

Enno helped by providing comments along the way. A particular challenge in the text arose from the fact that the work hasn’t been published yet, so Ekaterina had to remove some passages from the first draft – she didn’t want to reveal anything that might interfere with getting her article into a good journal.

It’s always possible to do more with a text; there are still passages that could be fuller, simpler, and clearer. But that’s the way it always is. As Mark Twain said, “The time to begin writing an article is when you have finished it to your satisfaction.” Somewhere he also stated, I believe, something to the effect that publication is the only way to force a writer to stop editing his piece.

(Mark Twain also advised writers to “Substitute ‘damn’ every time you’re inclined to write ‘very’; your editor will delete it and the writing will be just as it should be,” but perhaps that doesn’t apply here.)

So thanks, Ekaterina and Enno!


I’ve used the following structure: first comes Ekaterina’s complete original text (except for the passages that were removed). As you’re reading the original, make a note of anything that you think might cause problems for her target audience.

After the full original text I go it paragraph by paragraph, identifying specific issues that came up and showing how Ekaterina addressed them in her next draft.

That new version is assembled at the end. If you’re still alive at that point, it’s interesting to read the two side-by-side.


 

FIRST VERSION

An A-kinase anchoring protein regulates metabolism and motility in cancer cells

The ability of cancer cells to detach themselves from the initial tumor site, migrate and invade adjacent tissues signifies the first step of usually fatal distant metastasis. In order to achieve this, a complex organization of multiple intracellular processes initiates. Collectively, this transition is termed EMT (Epithelial-Mesenchymal Transition) and the ultimate goal of modern cancer research is to gain further insight into EMT induction and regulation.

While breaking bonds between cancer cells and initial tumor predicts poor prognosis, many proteins inside the cell require physical attachment to other proteins in order to function. Also, certain proteins have different roles depending on their location inside the cell and the switch between these roles occurs when the protein attaches itself to one cellular structure or to another. These kinds of attachments are usually facilitated by scaffolding proteins.

One example of location-dependent protein is GSK3β. When GSK3β is in the cytoplasm of the cell or in the nucleus, it has a main role in transferring biochemical information important for cells’ development as well as EMT. This information circuit is called the Wnt signaling pathway and it has been reported to malfunction in various tumors. However, when GSK3β in located in the cellular structures called the mitochondria, it transmits information required for survival or death of abnormal cells.

In addition to being location-dependent, GSK3β activity depends on physical interaction with other proteins. Under normal conditions GSK3β remains active and can be inhibited by a phosphate molecule placed on it by one of the proteins called kinases. One example of such a kinase protein is Protein Kinase A (PKA). The scaffolding protein in charge of bringing GSK3β and PKA in close proximity to each other in order to insure the transfer of the phosphate molecule from PKA to GSK3β belongs to the A-kinase anchoring proteins (AKAPs) family. AKAPs have in common the ability to bind PKA and direct it to a particular location inside the cell. Since this AKAP is able to regulate GSK3β activity via PKA by direct binding of the two proteins, elucidating the AKAP’s exact function in tumorigenesis is compelling.

Interestingly enough, when we delete the AKAP gene in a model cell line, thereby preventing the formation of the AKAP protein, …

(Here a section has been removed)

Reprogramming of cellular metabolism is another hallmark of cancer. Back in 1929, Otto Warburg first noted that rapidly growing cancer cells rely on different energy production methods than healthy slow-growing cells. He stated that cancer cells produced most of their energy in a process called anaerobic glycolysis, in contrast to healthy cells which preferred the highly-efficient energy production via oxidative phosphorylation (OXPHOS) which takes place in the mitochondria. The advantage of glycolysis over OXPHOS is the high speed in which energy is produced. While Warburg thought that all cancer cells switch to high-rate glycolysis, recent studies have shown that this is not entirely true; only rapidly growing and dividing cancer cells use glycolysis, while non-dividing or metastasizing cancer cells, prefer OXPHOS.

(Section removed)

In summary, this research demonstrated that the AKAP is required for both migration and metabolic control of lung cancer cells. However, the direct target protein of the AKAP remains to be found. Therefore, future work will aim to answer the “chicken or the egg” question; who was affected first by the AKAP, the EMT or metabolism?

Prospectively, our approach contributes to elucidating mechanisms underlying EMT and metabolic reprograming regulation in tumorigenesis and proposes this AKAP as a potential therapeutic target.


 

COMMENTS

In this section I walk through the text and comment on each issue as it arises. In classroom teaching it’s better to cluster problems of the same type and cover several examples before moving on to each new issue. This helps a student decontextualizes a problem and recognize it in their own writing. Here, for the sake of simplicity, I’ll stick to the order of the text. I’ll break it into paragraphs to discuss the overall information structure and transitions, then work through the sentence level to cover issues there.

First paragraph:

The ability of cancer cells to detach themselves from the initial tumor site, migrate and invade adjacent tissues signifies the first step of usually fatal distant metastasis. In order to achieve this, a complex organization of multiple intracellular processes initiates. Collectively, this transition is termed EMT (Epithelial-Mesenchymal Transition) and the ultimate goal of modern cancer research is to gain further insight into EMT induction and regulation.

 

I found this beginning logical; it starts with a concept that most readers are probably familiar with (“metastasis”) and moves to something that is probably new (EMT) that is crucial to the paper. Some of the sentences are dense, partly because they use words that are familiar but rather uncommon. If you can replace such words with more common terms without losing the meaning, it requires less “processing” by the reader. In a single sentence this may not be too important, but using such words constantly in a text makes it “heavier” and a bit tougher going overall.

In my own writing, I aim to compose sentences that make a point as clearly as possible, that can be understood the first time they are read, and whose structure links to the sentences that precede and follow them.

The first sentence will probably be accessible to the target audience (students and others who have no specialist knowledge of the field). It isn’t the easiest type of sentence for readers to decode because the subject (a list) goes on for a line and a half before we get the verb. There are several ways to avoid this:

Metastasis, which is usually the fatal stage in cancer, begins when...

Cancer cells take the first step toward metastasis… when they…

The sentence also contains several words and constructions that have alternates in more typical daily language: “The ability of… detach themselves (free themselves, break away from)… initial tumor site (the tissue where a cancer originally appears)… signifies (is)…” I wasn’t too concerned about this, but the phrase “usually fatal distant metastasis” compresses several ideas into a compound that is looser and will surely be easier to cope with after Ekaterina changed it to:

The ability of cancer cells to detach themselves from an initial tumor site, migrate and invade other tissues signifies the first step of metastasis, which is often the fatal step in cancer progression.

It’s always good to take out extra information if it doesn’t really contribute to the point at hand; here she has removed “adjacent” and “distant” and loosened the dense compound at the end of the first sentence by starting with a single, familiar word (metastasis) and adds a clause to provide the clarification or definition.

The second sentence begins with, “in order to achieve this,” which has unnecessary words (“To achieve this” would mean the same thing). “This” is vague because it refers to a complex process with many parts. “A complex organization of multiple intracellular processes” will be abstract to most readers, and it is dense because it compresses a sentence into a noun phrase. An alternative would be something like, “A number of complex processes within cells must be reorganized…” In the original version, again we have to wait quite a while for the verb (“initiates”), and most readers will find the sentence easier to scan with a shorter subject placed closer to its verb.

Here’s her solution:

Cells must reorganize a number of internal processes to initiate metastasis.

The next sentence starts with “collectively,” a word that’s common in scientific texts and a little more unusual in other styles. Its meaning should be clear, so we’ll leave it. But the reader must completely understand “what is being collected,” and there are ways to make this easier to scan.

I was worried that “the ultimate goal of modern cancer research” might be overstating her case: cancer is a vast field, whose ultimate goal is probably to cure the disease. EMT is surely a crucial step along the way, but other labs working on other aspects of cancer might have other “ultimate goals” and disagree. In her final version, she has toned this down.

“Initiate” and “regulate” are less-familiar than more common words such as “start” and “is controlled”, but it’s important to note that “regulation” has a particular scientific meaning. This is similar enough to the way most people understand the word that she might not necessarily want to change it. Here is her final version:

The ability of cancer cells to detach themselves from an initial tumor site, migrate and invade other tissues signifies the first stage of metastasis, which is often the fatal step in cancer progression. Cells must reorganize a number of internal processes to initiate metastasis. Collectively, the steps in this process are called the Epithelial-Mesenchymal Transition (EMT). An important goal of modern cancer research is to learn more about the molecules and processes that initiate EMT and influence the way it develops in a tissue.

 

Second paragraph:

While breaking bonds between cancer cells and initial tumor predicts poor prognosis, many proteins inside the cell require physical attachment to other proteins in order to function. Also, certain proteins have different roles depending on their location inside the cell and the switch between these roles occurs when the protein attaches itself to one cellular structure or to another. These kinds of attachments are usually facilitated by scaffolding proteins.

The connection between the introductory paragraph and this one is a crucial step along the way to Ekaterina’s complete, logical story. Her strategy raises an issue that often appears in writing about biomedical themes. When the experiments are finished and the results interpreted, it’s often possible to link something you have learned to cancer or another disease. That may have been the original intent of a project, or the connection may turn up while you’re pursuing some other question.

Either way, the claim will only be taken seriously if specific types of evidence are provided. In the type of work carried out at the MDC, this usually starts at a very basic level of a biological system, identifying a molecule, studying its activity in healthy cells, demonstrating what happens if it is missing or unable to perform its functions, and showing that the same type of disruption occurs in a disease.

This provides a logical framework to explain what you did – providing you approach things from the bottom up, from the lowest level of structure in an organism to one of the highest levels (health). But if your text starts with a disease, your entry point is high, and it’s harder to build the logic this way.

There are reasons to do so anyway: In popular science writing, diseases presumably attract readers out of a sort of diffuse self-interest: people are afraid of diseases and interested in progress toward cures. In scientific texts, diseases are often hung over a project like a big banner, perhaps to attract the attention of a wide community of cancer researchers, or to open the coffers of agencies that prefer to fund projects that will lead to real medical applications.

Whatever your reason, starting with the disease can introduce a structural problem in a story that is more logical when told the other way around. Ekaterina needs to jump from tumors and metastases to a scale that is millions of trillions of times smaller: interactions between proteins in cells. She wants to do that in the fewest steps possible so that she can start talking about her work. Her original draft uses a sort of logical slight-of-hand: she has introduced metastases and the notion that cancer cells have to detach themselves from their neighbors, and jumps to the idea that proteins have to attach themselves to structures inside cells. It’s awkward because she’s talking about things at vastly different scales, in different locations, and hasn’t provided any other logical connection between these events.

Ekaterina found a great solution to the problem by linking the two themes at a more profound level: cellular attachments and detachments depend on interactions between proteins – whether they bind to each other or not. Adding that connection makes the transition to the molecular scale much more logical:

Before a cancer cell can migrate away from its tumor, it has to detach itself by breaking its bonds to other cells, and this is generally a sign that the patient’s prognosis will be poor. The cell is usually tied to its neighbors and material in the space between them by proteins that are bound to each other; now these connections must be broken. The binding and unlinking of proteins, inside the cell as well as outside, is a general phenomenon that is crucial to every aspect of cellular life and has to be carefully controlled. Whether a protein binds to the right partner, and when and where it does so, can make the difference between life and death for a single cell or an entire organism.

A molecule’s location influences its ability to carry out its tasks, and its location is determined by interactions with other proteins that help attach it to a cellular structure or compartment. This often requires the participation of a “scaffolding” molecule which can bind to both the protein and the target membrane or structure it should be attached to; the scaffold provides a way of bringing them together.

Third paragraph:

One example of a location-dependent protein is GSK3β. When GSK3β is in the cytoplasm of the cell or in the nucleus, it has a main role in transferring biochemical information important for cells’ development as well as EMT. This information circuit is called the Wnt signaling pathway and it has been reported to malfunction in various tumors. However, when GSK3β in located in the cellular structures called the mitochondria, it transmits information required for survival or death of abnormal cells.

Here Ekaterina uses the concept of localization as a bridge between paragraphs; she just mentioned it, so readers should be able to follow her logic. Here I was concerned about the number of ideas that might be new which she introduced, and whether she had connected them to each other in the clearest possible way. Sometimes the simplest things make a text challenging to readers unfamiliar with basic concepts: What do they imagine when they read “transferring biochemical information,” and will they instantly relate it to the “information circuit” in the next sentence? The author can help by making connections excruciatingly explicit; readers who can’t follow your logic will have to invent one on their own, and rather than invest the effort, they may give up.

There are countless ways to build connectivity and make transitions. A lot of them don’t need to be as explicit when writing for scientists from your field. They are already familiar with concepts such as protein-protein binding and the logic that connects this theme to events at the scale of cells, including metastases.

Even in this type of communication, however, the writer has to be intensely aware of the logic that connects ideas to each other and the larger story; at that point you can make an intelligent decision about each transition. Is a connection really implicit, from the text? Would there be anything wrong with making it explicit? When giving someone directions to a party, it’s usually better to give too many directions than too few – within limits. If everyone you’re inviting lives in town and is familiar with the same landmarks nearby, you won’t need to start with a map of contintental Europe. In the same way, a scientist doesn’t need to be told that your body is made of cells.

As always, the amount of information and logic you choose to present should be guided by a realistic guess about the knowledge of your intended audience. Ekaterina added more linkage and rearranged some of the points within sentences to produce this version:

One example of a protein whose activity depends on its location is a molecule called GSK3β, which has been found to malfunction in various tumors. When GSK3β is in the cytoplasm or the nucleus, its main role is to transmit information needed during cells’ development and for the regulation of EMT. In these two locations, the main function of GSK3β is mainly to send signals along a molecular “information circuit” called the Wnt signaling pathway. But GSK3β can also be found within other cellular structures called mitochondria. There it transmits information along a different route, with different effects: the signal helps determine whether abnormal cells survive or die. Both locations and signals play a role in whether cancer cells grow, survive, and metastasize. They may only be able to do so by interrupting the signals that GSK3β normally transmits to other proteins. Therefore, ensuring that GSK3β is in the proper location and functions correctly is probably crucial in cancer prevention and treatment.

 

Fourth paragraph:

In addition to being location-dependent, GSK3β activity depends on physical interaction with other proteins. Under normal conditions GSK3β remains active and can be inhibited by a phosphate molecule placed on it by one of the proteins called kinases. One example of such a kinase protein is Protein Kinase A (PKA). The scaffolding protein in charge of bringing GSK3β and PKA in close proximity to each other in order to insure the transfer of the phosphate molecule from PKA to GSK3β is a member of the family of A-kinase anchoring proteins (AKAPs).

Here there are clear links to the previous paragraph (GSK3β and localization) and Ekaterina connects this to a topic introduced earlier: protein-protein binding. But within the paragraph, it’s not always immediately clear how each new idea arises from the previous one – at least until you have read the whole sentence. The first sentence, for example, ends with protein interactions, a theme also taken up in the next sentence – but only after the introduction of a new idea (phosphorylation), and the link between the two ideas comes at the end. The reader has to wait for the author to close the gap. The alternative is simply to change the order of information within sentences to get a better flow.

English permits great flexibility in the arrangement of phrases within sentences. Word order can be used to establish (or at least support) the logical flow and to transmit other kinds of meaning. Each sentence is like a story that begins somewhere and takes us somwhere else, and the next sentence can start right at that point and move on.

That is much more difficult in a language like German, where sentence structure is much more rigid: putting almost anything before a subject requires an inversion that pushes the main verb all the way to the end – so you often have to process entire sentences to grasp their connections. German readers are used to this and may, fundamentally, be better at it.

Ekaterina’s native language is not German, but she has used this type of sentence structure. To represent the problem symbolically, an arrangement of ideas is probably easiest to follow if it looks something like this:

A – B; B – C; C – D… etc.

If a sentence contains three pieces of information as Ekaterina’s does, this may be harder. But the arrangement in the first two sentences look more like this:

A – B – C. D – B – A

(A = localization. B = activity. C = protein interactions. D = phosphorylation. B = activity. A = interactions)

Rearranging this to create a more linear storytelling structure would change this:

In addition to being location-dependent, GSK3β activity depends on physical interaction with other proteins. Under normal conditions GSK3β remains active and can be inhibited by a phosphate molecule placed on it by one of the proteins called kinases.

….to something like this:

GSK3β’s activity depends on both its cellular location and its direct physical interactions with other proteins. By binding to a protein called a kinase, for example, it acquires a chemical tag called a phosphate group. This has an important fact: it inactivates GSK3β and switches off signals when a message has been received.

It’s not easy to carry this approach through to the end of the paragraph, because most of the sentences convey more than two ideas. That means shuffling and combining them in other ways. And a linear structure isn’t always appropriate: it’s good when describing a sequence of events, or when “zooming in” from a general idea to a specific one. Sentences that introduce lists, however, would be structured differently.

Suppose Ekaterina had introduced the topic this way: “In different cellular locations, GSK3β interacts with different sets of proteins and has different functions.” If examples follow, the clearest structure would be to construct the information the same way: “In the cell nucleus, GSK3β binds to specific proteins as a way of transmitting signals… In the cytoplasm it functions the same way. But when it is located in structures called mitochondria, …”

Ekaterina’s found a different solution that has the same effect: within sentences, the order of ideas reflects the logic that connects them:

GSK3β’s location and activity are the result of interactions with other proteins. When GSK3β binds to a molecule called Protein Kinase A (PKA), for example, PKA transfers a chemical tag called a phosphate group to it, which switches off the signaling activity of GSK3β. The tag can only be transferred if the two molecules are brought into direct contact, a process that is arranged by scaffolding proteins such as members of the family of A-kinase anchoring proteins (AKAPs).

This brings Ekaterina directly to the heart of her project. At this point the reader has been introduced to all the main players in the story, and now she is going to guide us to the specific question she is asking and the way she chose to pursue it.

 

Fifth paragraph:

AKAPs have in common the ability to bind PKA and direct it to a particular location inside the cell. Since this particular AKAP is able to regulate GSK3β activity via PKA by direct binding of the two proteins, elucidating the AKAP’s exact function in tumorigenesis is compelling.

Here there is another dense cluster in the first sentence: “A-kinase anchoring protein (AKAP) family,” which she will loosen in the final draft. In the second, “have in common the ability to” is awkward (and grammatically suspect because a phrase is interposed between the verb and its object).

The third sentence is interesting for another reason. It seems to fit the principles we have covered – it begins with points that have already been introduced and then builds a link to the overarching topic: the development of tumors and metastases. This raises a different issue related to information density. “Since the AKAP is able to regulate GSK3β activity via PKA by direct binding of the two proteins” combines several pieces of information that the reader has learned moments ago. That integration is important, but readers will only understand it if they have digested what they’ve learned. It’s a bit like introducing someone to three new words in a foreign language and a new grammar rule, and then immediately presenting him with a sentence that combines them all. He may understand it, but you shouldn’t expect him to.

Here’s how Ekaterina loosened it up:

AKAPs share a common feature: the ability to bind PKA and transport it to a particular location inside the cell. This means that the AKAP not only interacts with both GSK3β and PKA – it also directs them to specific locations in the cell. That’s interesting because PKA also has many roles in the development of tumors.

In the new draft she decided to expand on this by reminding readers of a point she had raised earlier:

Now three molecules are bound together: the AKAP, PKA and GSK3β. This puts PKA close enough to transfer a phosphate group onto GSK3β and block its activity. Since both of these molecules have been implicated in the development of tumors, it makes sense to wonder whether the AKAP, the protein that brings them together, might also have a role in the disease. If so, we would probably expect to find a disruption of the normal activity of the AKAP, but its functions in healthy cells have been unclear.

 

Sixth paragraph

At this point Ekaterina has introduced the main question of her research, which is something like this: “Are the cancerous effects of GSK3β sometimes related to the fact that it is in the wrong place and/or its ability to interact with important partners such as PKA? And if so, could defects in the AKAP be responsible?”

Her thesis project required breaking this question down into smaller parts, designing experiments to address each part, and then assembling the results into a satisfactory answer. In the sixth paragraph she introduces part of the experimental strategy:

Interestingly enough, when we delete the AKAP gene in a model cell line, thereby preventing the formation of the AKAP protein, …

She hasn’t explicitly explained the rationale for this experiment – which any scientist will understand, but what about non-specialists? There’s an easy way to make sure they get the point:

A common method to discover the function of a molecule is to remove it from cells where it is normally found and observe what happens to them. We did this with the AKAP by deleting its gene in a line of cells that we use as a model in the lab, which left the cells incapable of producing the AKAP protein. We discovered that its removal affects a number of processes that are specifically involved in metastasis and other aspects of the development of tumors.

Next Ekaterina confronts a technical problem: She can’t describe some of her experiments or present the results until they have been published in a scientific journal. Giving away too many details, even in the MDC newsletter, would be grounds for the rejection of her paper by a journal – which typically refuses to publish work that has already appeared.

Her solution is to address the reader directly and frankly, explaining why she doesn’t go deeper into the project:

For the details you’ll have to wait for the paper. For now we can say that the silencing of the AKAP affects the behavior of other proteins that play crucial roles in signaling, EMT, and also cell metabolism, the process by which cells produce the energy they need. Cancer cells have different energy needs than healthy cells, and the reprogramming of cellular metabolism is another hallmark of cancer.

Metabolism is another theme she wants to introduce because it has an important role in her thesis, and she finds a great way to do so using history:

Back in 1929, Otto Warburg first noted that rapidly growing cancer cells rely on different energy production methods than healthy slow-growing cells. He stated that cancer cells produced most of their energy in a process called anaerobic glycolysis, in contrast to healthy cells which preferred the highly-efficient energy production via oxidative phosphorylation (OXPHOS) which takes place in the mitochondria. The advantage of glycolysis over OXPHOS is the high speed in which energy is produced. While Warburg thought that all cancer cells switch to high-rate glycolysis, recent studies have shown that this is not entirely true; only rapidly growing and dividing cancer cells use glycolysis, while non-dividing or metastasizing cancer cells, prefer OXPHOS.

This is a good storytelling because it gives readers a character to latch onto. By placing her project in a historical context, she is also saying more, by implication: her work addresses an issue so fundamental in cancer research that it has occupied scientists for nearly a century.

All along the way Ekaterina confronts important decisions about the amount of information a reader really needs to get the gist of her work. Here she decides to introduce two types of metabolism that she names without going into much detail; it’s enough to contrast them based on the sites in cells where they occur and a parameter that is related to cancer (speed). I recommended sharpening the contrast by expanding on the idea in this sentence:

The advantage of glycolysis over OXPHOS is the high rate at which energy is produced.

Here’s what she came up with:

This link was discovered back in 1929, when the German scientist Otto Warburg became the first to note that rapidly growing cancer cells rely on methods of energy production that are different than those used by healthy, slow-growing cells. He stated that cancer cells produced most of their energy in a process called anaerobic glycolysis. Healthy cells, on the other hand, favored a type of energy production based on a process called oxidative phosphorylation (OXPHOS), which takes place in the mitochondria. The difference has to do with speed and efficiency: glycolysis produces energy at a very high rate, while OXPHOS is geared toward efficiency, and is more sustainable over the long term. While Warburg thought that all cancer cells switch to high-rate glycolysis, recent studies have shown that this is not entirely true; only rapidly growing and dividing cancer cells use glycolysis. Non-dividing or metastasizing cancer cells prefer OXPHOS. When you read the paper you’ll see how we demonstrated the AKAP’s effects on metabolism and how we interpret them in terms of the dysregulation of metabolism and EMT.

(Section removed)

The reader has to guess how this might relate to her project, but some clues are available: OXPHOS takes place in the mitochondria, which is one of the sites where GSK3β is found… Hmm…

Now all that’s left is to sum up. The conclusion is crucial because it gives the author a chance to put everything back together, to show how she has answered the scientific question posed at the beginning, and to frame the story she wants the reader to remember it. Here’s what she wrote for the first draft:

In summary, this research demonstrated that the AKAP is required for both migration and metabolic control of cancer cells. However, the direct target protein of the AKAP remains to be found. Therefore, future work will aim to answer the “chicken or the egg” question; who was affected first by the AKAP, EMT or metabolism?

Prospectively, our approach contributes to elucidating mechanisms underlying EMT and metabolic reprograming regulation in tumorigenesis and proposes the AKAP as a potential therapeutic target.

The main changes she ends up making have to do with phrasing, word order, and making sure her logic (which would be clear to any scientist) will also make sense to readers less familiar with buzzwords like “elucidating mechanisms” and “potential therapeutic target.” She gets rid of clusters like “contributes to elucidating mechanisms underlying EMT and metabolic reprogramming regulation in tumorigenesis.” By now most of those terms will have confronted the reader, but the conclusion is the last place you want the text to get dense.

A lot of questions remain: other proteins directly affected by the AKAP have yet to be found. The project raises a “chicken-or-egg” question, in which EMT is the chicken and metabolism the egg: which process does the AKAP influence first? Our approach should contribute to understanding the mechanisms that produce some of the changes in these processes that are observed as tumors arise from healthy tissue and then become metastatic. And it hints that the AKAP might make a therapeutic target. This would be useful because AKAPs have features that might permit fine-tuning their effects with drugs. By doing so, it might be possible to alter the behavior of GSK3β in one location, where its activities contribute to disease, without affecting its healthy functions.

Below I provide the final version. As Ekaterina lives with this text, she’ll find that she could come back time and time again and improve it with editing. It’s a writer’s curse – to recognize weaknesses in old texts that have been published and you can no longer change. The important thing is what you learn when you mak the effort.


 

FINAL VERSION

Can a small A-kinase anchoring protein play a big role in cancer? A report from the lab bench

The ability of cancer cells to detach themselves from an initial tumor site, migrate and invade other tissues signifies the first stage of metastasis, which is often the fatal step in cancer progression. Cells must reorganize a number of internal processes to initiate metastasis. Collectively, the steps in this process are called the Epithelial-Mesenchymal Transition (EMT). An important goal of modern cancer research is to learn more about the molecules and processes that initiate EMT and influence the way it develops in a tissue.

Before a cancer cell can migrate away from its tumor, it has to detach itself by breaking its bonds to other cells, and this is generally a sign that the patient’s prognosis will be poor. The cell is usually tied to its neighbors and material in the space between them by proteins that are bound to each other; now these connections must be broken. The binding and unlinking of proteins, inside the cell as well as outside, is a general phenomenon that is crucial to every aspect of cellular life and has to be carefully controlled. Whether a protein binds to the right partner, and when and where it does so, can make the difference between life and death for a single cell or an entire organism.

A molecule’s location influences its ability to carry out its tasks, and its location is determined by interactions with other proteins that help attach it to a cellular structure or compartment. This often requires the participation of a “scaffolding” molecule which can bind to both the protein and the target membrane or structure it should be attached to; the scaffold provides a way of bringing them together.

One example of a protein whose activity depends on its location is a molecule called GSK3β, which has been found to malfunction in various tumors. When GSK3β is in the cytoplasm or the nucleus, its main role is to transmit information needed during cells’ development and for the regulation of EMT. In these two locations, the main function of GSK3β is mainly to send signals along a molecular “information circuit” called the Wnt signaling pathway. But GSK3β can also be found within other cellular structures called mitochondria. There it transmits information along a different route, with different effects: the signal helps determine whether abnormal cells survive or die. Both locations and signals play a role in whether cancer cells grow, survive, and metastasize. They may only be able to do so by interrupting the signals that GSK3β normally transmits to other proteins. Therefore, ensuring that GSK3β is in the proper location and functions correctly is probably crucial in cancer prevention and treatment.

GSK3β’s location and activity are the result of interactions with other proteins. When GSK3β binds to a molecule called Protein Kinase A (PKA), for example, PKA transfers a chemical tag called a phosphate group to it, which switches off the signaling activity of GSK3β. The tag can only be transferred if the two molecules are brought into direct contact, a process that is arranged by scaffolding proteins such as the members of the family of A-kinase anchoring proteins (AKAPs).

AKAPs share a common feature: the ability to bind PKA and transport it to a particular location inside the cell. This means that the AKAP not only interacts with both GSK3β and PKA – it also directs them to specific locations. That’s interesting because PKA also has many roles in the development of tumors.

Now three molecules are bound together: the AKAP, PKA and GSK3β. This puts PKA close enough to transfer a phosphate group onto GSK3β and block its activity. Since both of these molecules have been implicated in the development of tumors, it makes sense to wonder whether AKAP, the protein that brings them together, might also have a role in the disease. If so, we would probably expect to find a disruption of the normal activity of the AKAP, but its functions in healthy cells have been unclear.

A common method to discover the function of a molecule is to remove it from cells where it is normally found and observe what happens to them. We did this with the AKAP by deleting its gene in a line of cells that we use as a model in the lab, which left the cells incapable of producing the AKAP protein. We discovered that its removal affects a number of processes that are specifically involved in metastasis and other aspects of the development of tumors.

For the details you’ll have to wait for the paper. For now we can say that the silencing of the AKAP affects the behavior of other proteins that play crucial roles in signaling, EMT, and also cell metabolism, the process by which cells produce the energy they need. Cancer cells have different energy needs than healthy cells, and the reprogramming of cellular metabolism is another hallmark of cancer.

This link was discovered back in 1929, when the German scientist Otto Warburg became the first to note that rapidly growing cancer cells rely on methods of energy production that are different than those used by healthy, slow-growing cells. He stated that cancer cells produced most of their energy in a process called anaerobic glycolysis. Healthy cells, on the other hand, favored a type of energy production based on a process called oxidative phosphorylation (OXPHOS), which takes place in the mitochondria. The difference has to do with speed and efficiency: glycolysis produces energy at a very high rate, while OXPHOS is geared toward efficiency, and is more sustainable over the long term. While Warburg thought that all cancer cells switch to high-rate glycolysis, recent studies have shown that this is not entirely true; only rapidly growing and dividing cancer cells use glycolysis. Non-dividing or metastasizing cancer cells prefer OXPHOS. When you read the paper you’ll see how we demonstrated the AKAP’s effects on metabolism and how we interpret them in terms of the dysregulation of metabolism and EMT.

A lot of questions remain: other proteins directly affected by the AKAP have yet to be found. The project raises a “chicken-or-egg” question, in which EMT is the chicken and metabolism the egg: which process does the AKAP influence first? Our approach should contribute to understanding the mechanisms that produce some of the changes in these processes that are observed as tumors arise from healthy tissue and then become metastatic. And it hints that the AKAP might make a therapeutic target. This would be useful because AKAPs have features that might permit fine-tuning their effects with drugs. By doing so, it might be possible to alter the behavior of GSK3β in one location, where its activities contribute to disease, without affecting its healthy functions.

 

Advertisements