Passing the smell test

A couple years ago, Dr. Rita P.-Y. Chen published in Scientific Reports. Although her paper was called “Polyhydroxycurcuminoids but not curcumin upregulate neprilysin and can be applied to the prevention of Alzheimer’s disease,” her results started showing up in the popular press as, “Eat Indian food to prevent Alzheimer’s disease!” Her experience became a good lesson for my students as we practiced limiting jargon and providing interesting information.

A recent paper in Nature reminded me of the curry mania, as Dr. Gregory J. Hannon’s paper “Asparagine bioavailability governs metastasis in a model of breast cancer” was summarized in the Mirror as, “Asparagus causes breast cancer, but don’t worry, because they only tested this in mice.” I won’t even link to the Mirror, because the lack of detail and citation is alarming. For a good summary and some insightful commentary, check out “Spread of breast cancer linked to compound in asparagus and other foods” in The Guardian:

Asparagine is an amino acid that is made naturally in the body as a building block for proteins. But it is also found in the diet, and in high levels in certain meats, vegetables and dairy products.

The international team of cancer specialists from Britain, the US, and Canada studied mice with an aggressive form of breast cancer. The mice develop secondary tumours in a matter of weeks and tend to die from the disease within months.

Writing in the journal Nature, the researchers describe how they reduced the ability of breast cancer to spread in the animals by blocking asparagine with a drug called L-asparaginase. To a lesser extent, by putting the animals on a low-asparagine diet worked too. Inspired by the results, the scientists examined records from human cancers and found that breast tumours that churned out the most asparagine were most likely to spread, leading patients to die sooner. The same was seen in cancers of the head, neck and kidney.

I highly recommend the article from The Guardian (and supporting quality journalism) and won’t recapitulate the main points here. Instead, let’s talk a bit about metastasis, or “How did they find breast cancer in my mother-in-law’s femur?”

Metastasis means movement–cancer cells that originated in breast tissue moved into the bloodstream, traveled to a different organ (like the lungs or brain), and were able to generate more cancer cells in the new location. Most cells stay in one place, so scientists looked into the differences between cells that stay and cells that move. (In fact, most cells don’t become cancerous, so scientists also look into the differences between normal and cancerous cells.) In this study, one of the ways the “stay” cells differed from the “move” cells was how much asparagine they could produce. “Move” cells produced a lot more asparagine than the “stay” cells.

That’s only half the evidence.

The other half comes from what happens when the scientists reduced how much asparagine was in the cells–the “move” cells weren’t able to move.

So, WHY doesn’t this study mean that we should all stop eating asparagus? Hey, I’m not a huge fan of asparagus, so don’t let me stop your boycott, but don’t do it because you think you’ll prevent breast cancer. For one thing, asparagine is an amino acid that is in a lot of food, not just asparagus. For another, this study is talking about cancer cells that stay versus cancer cells that move. Their conclusions don’t necessarily mean that eliminating asparagine can prevent cancer altogether.

There are some cool findings in this paper, but simply claiming (AHEM, MIRROR) that asparagus causes breast cancer does NOT pass the smell test.

PS. Did you think that the “smell test” had to do with asparagus making your urine smell bad? You can read more about that at the Smithsonian (from 2013) and at NPR (from 2016).


All about that cake

Last year, a friend asked me how I felt about moon cakes. As it turns out, I have a LOT of feelings about moon cakes (and other Taiwanese “cakes”). This evolved into a collaboration with the talented Julia Kuo, and I’m pleased to share our work, published recently over at The Cleaver Quarterly:

Cake in Taiwan

Mid-Autumn Festival falls on the biggest full moon of the year, and it also happens to be my lunar birthday. Every year, I celebrate twice: butter/oil cake for my Gregorian birthday, and mooncakes filled with red bean paste or pineapple for my lunar birthday. With the vast selection of mooncakes in the days leading up to Mid-Autumn Festival, I could choose a “Taiwanese birthday cake” as big as 4 inches in diameter, but I prefer the smaller mooncakes (ice-cream scoop-sized) for the perfect filling-to-flaky-crust ratio.

Go check out the gifs!

Let them eat cake! A brain signal that delivers a one-two punch to obesity and cancer

Healthy eating and exercise are doctors’ top two recommendations to treat a wide range of ailments. Unfortunately, New Years’ resolutions for fitness often find their way to an early February grave, and we dismiss these lifestyle changes that could save our lives. The barriers to exercise involve both time and access, and the convenience of unhealthy food often outweighs a balanced diet. The latter was dramatically portrayed in the documentary Super Size Me, which followed one man’s monthlong descent into sedentary binge-eating. While diet and exercise are suggestions at the macro level, there is a molecular basis to physician-recommended changes. Scientists have found a molecule that links functions in the brain to obesity, diabetes, and cancer. This molecule, brain-derived neurotrophic factor (BDNF), is a signal from the part of the brain that regulates body mass. As a result, varying BDNF levels dramatically impacts overall health.

Studying lifestyle changes in mice

Researchers already know that obesity is one of the leading causes of breast cancer, and the Ohio State University research group led by Dr. Lei Cao realized that BDNF could be the culprit when they noticed that improving living conditions lowered obesity and cancer risk in mice.

Researcher Grant Foglesong described the improved conditions to include “increased living space that could house more mice, leading to a more advanced social hierarchy.” In addition to stimulation from social interactions, “The mice also had toys to play with, like tunnels and running wheels, which normal lab mice don’t get.”

The mice living in this improved environment had higher levels of BDNF in the hypothalamus, which is the part of the brain that controls body mass. This led researchers to ask whether increasing BDNF in mice living in standard lab conditions would lower their susceptibility to obesity and cancer without changing their lifestyles.

A new fate for Supersized mice

Similarly to filmmaker Morgan Spurlock eating three McDonalds meals per day and not exercising, the lab mice were fed a fatty diet without access to exercise wheels. After one year, the mice had become morbidly obese and diabetic. When researchers injected extra BDNF into the hypothalamus, the mice became noticeably healthier, even though their diet and activity level were unchanged.

With increased BDNF, the obese mice returned to a healthy body weight. Although they continued to eat the fatty diet, after ten weeks they were able to reach and maintain a healthy body weight. Alongside the reduction in fat tissue, researchers saw improved health in the mice in terms of other obesity-related diseases.

Type II diabetes, where patients still produce insulin, but their bodies can no longer use it, is one disease related to obesity. Because obese mice lack proper insulin response, they typically have an exaggerated spike in blood sugar after a glucose injection. The mice treated with BDNF were better able to manage a sudden glucose spike, showing that their bodies could now respond to insulin.

After reversing obesity-induced diabetes using BDNF, the Cao group turned to the link between obesity and cancer. Prognosis in cancer patients is typically worsened in obese individuals. However, the obese mice treated with BDNF were able to slow tumor growth after an injection of breast cancer cells. These promising results will drive research for a possible therapy to treat both obesity and breast cancer. 

So can we have our cake and eat it too?

Following a regular exercise regimen and maintaining a balanced diet increases BDNF levels in humans. So, can we skip the healthy lifestyle and just inject ourselves with BDNF?

“Well no,” Foglesong cautioned. “Increasing BDNF levels in human patients would involve brain operations that could prove risky.”

Still, there is hope that being able to target the brain-cancer connection will mitigate the difficulties in following medical advice. “If we can develop a less invasive BDNF treatment, we will be able to reverse obesity and potentially prevent cancer.”

***This was based on an article published in Molecular Therapy and an interview with (soon-to-be Dr.) Grant Foglesong.

Mol Ther. 2014 Jul;22(7):1275-84. doi: 10.1038/mt.2014.45. Epub 2014 Mar 18.

Hypothalamic gene transfer of BDNF inhibits breast cancer progression and metastasis in middle age obese mice.

The story of a year

Today marks one year of my stay in Taiwan. I pondered the relative merits of telling a quantitative story or a qualitative one, but then I realized there’s no need to choose. In addition to celebrating this year of family, friends, and pups, I hope to get back into blogging regularly (time to gear up for #BCMonth) and try my hand at making infographics!

Part 1: Dwelling

Year One

One of the first things I did after cabbing from the airport to my cousin’s house was to view an apartment. My relatives were kind enough to do some footwork for me beforehand, so I signed a lease for the second apartment (first building) I viewed on Day 1! A year later, I’ve been able to add a bit to the decor by hanging up pictures and postcards and bringing in a coffee table and some rugs.

It’s been great to have (a tiny bit of) space for visitors, and I’m thankful for everyone who has come to visit from all around the world. Who wants to stop by next?

#BCMonth Roundup with ALL THE LINKS

Well, we are reaching the end of October. I’m a bit disappointed that I didn’t post every single day, so I think I will continue this series beyond this month. In the meantime, let me round up the posts I did write this month:

Never fear–this isn’t the end of my breast cancer posts (and since it’s October 30, it’s not even really the end of #BCMonth). In fact, I have at least ONE exciting post coming up (more if I find some suitably inspirational cake, right?). I hope you all enjoy your Halloween; may the ghouls and goblins be scarier than facing down cancer.

#BCMonth Breast cancer in men

Over 2,000 men are diagnosed with breast cancer each year in the United States. Because people don’t generally associate breast cancer with men, the disease might not be caught as early in men, which in turn affects treatment success and survival rates. The American Journal of Men’s Health published this paper by S. Hesse-Biber and C An about the medical decisions men make when they carry BRCA mutations. Although I’m not able to access the full paper, the abstract states that 70% of the respondents got tested because they were considering family risk; 22%, considering medical implications; 8%, social support.

Only 8% of these men thought they would get social support from knowing their BRCA status! Contrast this with the pink message and abundant reminders for women to perform self-checks and ask for cancer screening. I have a problem with social slacktivism, and thinking awareness is enough, but this is definitely a case where there isn’t enough awareness of the problem. Breast cancer IS a disease in men, and knowing about the possibility (particularly knowing genetic risk) will increase diagnosis and survival!

Interestingly, a group headed by Yael Laitman of the Chaim Sheba Medical Center reported in Breast Cancer Research and Treatment that male Jewish carriers of BRCA mutations have an increased risk of breast cancer and pancreatic cancer, but not prostate cancer, compared to those who do not have BRCA mutations. I was a bit surprised to see this, as BRCA defects are known to be involved in prostate cancer, and this group, headed by Henriette Roed Nielsen of Vejle Hospital, reported in Acta Oncologica that there is a potential prostate cancer cluster in BRCA2.

I hope that we see more research on this matter.

#BCMonth Doing my homework

Sorry for the hiatus for the past couple days. I’ve been busy preparing for the workshops I’m leading…and well, sometimes I get distracted by donuts. BUT I’M BACK. And, I’ve been doing my homework. Today, I’ll be talking about a paper I just read by David C. Qian at Dartmouth, published this month in Human Molecular Genetics

David C. Qian, Jinyoung Byun, Younghun Han, Casey S. Greene, John K. Field,Rayjean J. Hung, Yonathan BrhaneJohn R. Mclaughlin, Gordon Fehringer,Maria Teresa Landi, Albert Rosenberger, Heike Bickeböller, Jyoti Malhotra,Angela Risch, Joachim Heinrich, David J. Hunter, Brian E. Henderson,Christopher A. Haiman, Fredrick R. Schumacher, Rosalind A. Eeles,Douglas F. Easton, Daniela Seminara, and Christopher I. Amos. Identification of shared and unique susceptibility pathways among cancers of the lung, breast, and prostate from genome-wide association studies and tissue-specific protein interactions. Hum. Mol. Genet. first published online October 19, 2015 doi:10.1093/hmg/ddv440

This is a good chance to talk about genome-wide association studies, or GWAS. (By the way, you pronounce that jee-wass. I know you were wondering.) The “genome-wide” part of that term means that researchers look at ALL THE GENES. They look for single-nucleotide polymorphisms, or SNPs (pronounced “snips”). A SNP is a base (remember, one rung on the DNA ladder) that differs among different people. What does that mean? Well, maybe on my ladder, the 7th rung is an A. My parents probably have the same. But, on my friend’s ladder, the 7th rung is a C (and her parents have a C, too). These are the sequence variations we’ve been talking about–they might not change any functions in the cell, but they might also have a HUGE impact. So, in a GWAS, researchers look at all the SNPs in two groups of people: individuals who have the disease/condition, and individuals who don’t.


The idea here is that SNPs you can find in the orange population might be the reason those individuals got cancer. Comparing the two groups helps researchers narrow down the field of genes that they test. The problem is that researchers don’t think we’re looking for a BRCAX anymore (a gene that would follow BRCA1 and BRCA2). Known BRCA mutations only account for about 20% of familial breast cancer, while other mutations account for even fewer cases. So, it’s important to look at pathways. One gene might not be responsible for someone getting cancer, but the other genes it interacts with might all work together–a crime syndicate, you might say.

For this specific paper, the researchers looked at three different cancers (and I color-coded them below with their “support ribbon” colors) to see what SNPs were present in all three kinds of cancer, and what SNPs seemed to be specific to one kind of cancer.


After comparing the different groups of SNPs, they looked at the genes were these SNPs were located, and analyzed the pathways that use these genes. There were some general pathways for growth and survival that were found in all three types of cancer, but each type of cancer also had specific pathways that stood out. In lung cancer, there were SNPs in genes involved in interleukin signaling, which is how transcription factors get turned on to tell cells, “HEY START MAKING THIS STUFF.” In breast cancer, there were SNPs in genes involved in apoptosis initiation,which is how cells end their life cycles (think about how your skin cells slough off daily). In prostate cancer, there were SNPs in genes involved in hypoxia response,which is important for cells that are buried inside tumors that don’t get enough oxygen.

This kind of study is important because it helps us understand how different types of cancers get started (and it might have to do with what these different tissues and organs are used for), and what targets we might want to use in drug development. I’m also curious about the similarities among the types of cancer. My advisor once quipped, “The prostate–it’s like the male breast!” He said this because a lot of the factors involved in breast cancer susceptibility also come into play in prostate cancer. It makes me wonder if there has been any tracking–do the men who get breast cancer (about 3,000 a year in the United States) have a higher rate of developing prostate cancer? I’ll have to see if I can find out.

#BCMonth New Guidelines!?

I’ve harped on early detection, but also highlighted physicians who caution against hasty surgery or chemotherapy. It’s important to balance vigilance with paranoia, and today’s news draws that contradiction into the spotlight once again. Karen Kaplan of the LA Times has written a thorough and informative article about the American Cancer Society’s new guidelines for mammography, so I’d like to direct you there to read more about it. Instead of recapitulating those points, I’d like to summarize some of the pros and cons, or why people argue over these guidelines suggesting fewer mammograms or starting them at a later age.


  • Mammograms have false positives, which might lead to more invasive tests that turn out to be unnecessary
  • Patients don’t have to spend as much money or time on the procedure
  • Individuals who don’t have a high risk of breast cancer don’t need to be screened as frequently


  • Patients whose cancer was caught early would rather go through all the tests than have the chance that their disease was missed
  • The false positive rate is tied to film mammography, when most mammograms in the United States are digital
  • Most individuals don’t know their risk

I think the last point is important, so let’s talk about it a bit more. How is someone supposed to know his or her risk? (And remember, men get breast cancer too!) Some people know to be extra vigilant because they have a family history of breast cancer. Some people have gotten their DNA sequenced to see if they have high-risk mutations. Well, in the past, that kind of sequencing wasn’t always accessible; many insurance companies wouldn’t pay for the test unless there was documented family history (not just of breast cancer, but that the breast cancer cases were caused by known mutations). The sequencing is more affordable now, but not everyone will get it done. Moreover, there are a lot of “inconclusive” results, which goes back to the variants of unknown significance that we’ve been talking about.

It’s important that we figure out what all these variants mean so that people and their physicians can make informed medical decisions; yearly screens might NOT be useful for everyone, but we need to make sure that the at-risk population knows how often to get them!

#BCMonth Gene Sequencing part II

Ok, we’ve talked about how double-stranded DNA can act as a template to copy DNA, but how does that connect to gene sequencing? Well, when the DNA gets copied, it’s not just presto-change-o, NEW DNA! The machine that does the copying slides along the template strand and adds a base at a time. Each base that it adds has a little “hook” for the next base to be added.


This keeps going as long as there’s a template and a supply of bases with little hooks.


For sequencing, researchers mix in bases that don’t have little hooks.


If one of these hook-less bases gets added, then the reaction stops there–nothing more gets added.


Just like there are long words and short words, genes can be long or short–with thousands of bases, or just a couple hundred. If there is a supply of bases with hooks, then the whole gene can be copied. Remember that this can happen pretty fast–one copy turns into two copies, which turn into four, eight, sixteen, and so on. Well, we can use that supply of bases but add in some A-bases that don’t have the hook. Every time there is a T on the template strand, the machinery adds an A, right? Well, sometimes the machinery grabs an A with a hook and keeps going, but sometimes, the machinery grabs an A without a hook, so the DNA copying stops there. This way, you end up with a bunch of fragments, and at the end of each fragment is an A. (In this picture, the As that don’t have a hook are purple.)


Say you get fragments that are 1 base, 5 bases, and 9 bases long. That means that the first “letter” of this “word” is A, and so are the fifth and ninth. Ok, make up a mix of normal bases, and this time, add in some Ts without hooks. The fragments you get from this mix will tell you where in the sequence there are Ts. You can do the same with Cs and Gs. Put all the clues together, and you might just get as lucky as this guy did:

#BCMonth Gene sequencing Part I

We’ve been talking a lot about sequencing, and I want to remind you to sign this petition to make data more accessible to patients and researchers.

You might ask–how do you sequence genes? So, let’s get into that today. Remember how DNA looks like a ladder? Well, DNA is double-stranded, so you can actually separate the two strands of the ladder. We’ve talked about how the different genes are like words in a play, and you might spell the words differently, right? So, each of these “half-rungs” is a “DNA letter,” which we call a base.


The DNA alphabet has four possible bases: A, T, C, and G. The cool thing about these bases is that every time you have an A, the base across from it is a T, and every time you have a C, the base across from it is a G. So, you can tell that two strands belong together if they have complementary sequences. In this picture, I’m color coding: A is red, so its complement T is green; C is blue, so its complement G is orange.


What this “complementary pair” system means is that if you know all the bases on one side of the ladder, you can fill in the blank and figure out what’s on the other side, too. This is useful when a cell is copying its DNA (getting ready to divide into daughter cells), and it’s useful for researchers, too. You can split up the two strands and use them as a template for the next copy–see a T on the template? Add an A on the new strand. See a C on the template? Add a G on the new strand. Now you have twice the amount of DNA you had before, and it’s the same sequence (or same spelling) as the original version. You can split these new strands, copy again, and you have four copies. Split and copy–eight!


How does this relate to sequencing a gene? Stay tuned!