Cure your ills with yogurt! ORLY?

I came across an article which prompted my BS meter to go into overdrive this morning. This is via a “Business Report” in the San Francisco Chronicle entitled “Antidepressant-infused yogurt with mail-order DNA.” Let’s look for the  wrong in this article!

First a summary: the article describes the claims of Tuur van Balen, a Belgian bio-engineer, who has been on the motivational speaking circuit and telling audiences that yogurt might be infused with antidepressant properties. His premise is that the bacteria that are used to make yogurt might have novel properties added to them in the laboratory, by adding new DNA sequences to their genomes. The recombinant bacteria then might produce useful chemicals while forming a delicious breakfast treat! The premise on its own is realistic, and currently there is a large number of on-the-market foods that contain genetically modified organisms among their ingredients. The example given in his presentation (available via Youtube) got my ire up.

His video presentation showed audience members how to engineer yogurt which produces antidepressants, specifically Prozac. This claim bugged me, because it didn’t seem plausible. I watched the video, and the implication from it was that a single DNA sequence of just over 800 nucleotides would be sufficient to accomplish this task, however in truth it would be far more difficult to do this for many compounds, and it turns out that it would be impossible to accomplish using Lactobacillus.

Prozac is a small molecule, which is thought to operate by inhibiting the reuptake of the neurotransmitter serotonin. Although the chemical structure of Prozac is not particularly complex, it is not something that could be synthesized biologically and involves over a dozen organic synthesis steps. Consequently, it is fallacious to claim that the introduction of a single DNA sequence (presumably encoding one enzyme) might be able to introduce a biochemical pathway that could generate this compound in vivo.  Now, bacteria can and have been engineered with a number of exogenous DNA sequences that add a novel biochemical pathway. I discussed one such example on this blog, where E. coli was engineered to produce the small molecule antibiotic erythromycin. It’s a lot of work, and takes a lot of careful planning, and is not likely a DIY project.

So let’s consider then an approach that might be more fruitful, and potentially more DIY than this proposal to “have yogurt bacteria make Prozac.” Fortunately for us, alternatives exist! One way to accomplish this is through a new-ish technology called RNA interference, which was the subject of the 2006 Nobel Prize for Medicine or Physiology to Andrew Fire and Craig Mello. With this technique, synthesized pieces of RNA are introduced into an organism, where they can bind to messenger RNAs made by the cell. This can have the effect of “silencing” that RNA, resulting in loss of expression of the gene. It is a temporary measure, and as the cell continues to grow, the “silenced” gene may start to be expressed again. The technique as originally used by Fire and Mello was used to study gene expression in the roundworm Caenorhabditis elegans.

Student researchers at York College are using these animals to study the effects of losing the function of specific genes. It is actually quite simple to do; the worms eat bacteria (E. coli) which have a DNA cassette added to their genome that encodes for the gene-specific  interfering RNA. The interfering RNA then enters the worm cells, and affects expression of the gene.

How could this be applied to humans? We could eat yogurt containing live cultures of recombinant, RNA interfering-expressing Lactobacillus. As indicated above, the drug Prozac acts to inhibit the internalization of the neurotransmitter serotonin at nerve synapses. This uptake process requires a membrane transporter that captures serotonin on the outside of the cell, and brings it inside the cell. Prozac presumably blocks the hole, preventing efficient uptake of serotonin. Instead, an interfering RNA might be introduced that decreases the level of the membrane transporter, accomplishing the same thing as introduced Prozac, but with only a single step!


About ycpmicro

My name is David Singleton, and I am an Associate Professor of Microbiology at York College of Pennsylvania. My main course is BIO230, a course taken by allied-health students at YCP. Views on this site are my own.

Posted on February 24, 2012, in Bogus!, You are what you eat and tagged , , . Bookmark the permalink. 4 Comments.

  1. I think that this is fascinating! From a psychology standpoint, I wonder if those patients who ate the yogurt that was not really an “anti-depressant” still felt better after they did. Sometimes (but certainly not in all situations) I feel that the mind is a very powerful tool..and when we are really convinced of something sometimes it does affect the physical symptoms we are experiencing. I also wonder if this new method of RNA interference will one day be advanced enough to help us overcome the resistance to anti-biotics we are encountering. Would it be possible to change the “resistance” expression of a gene in a bacterium?

    • Well, I for one would be wary of ingesting yogurt that can modulate my gene expression, so I am not planning on signing up for any clinical trials any time soon!

  2. After posting a question on our discussion board about pregnant women and AZT treatments, I noticed that there were a lot of “google posts” online about pregnant AIDS patients taking AZT’s. Because of our discussions in class about AZT affecting DNA synthesis, I began to wonder how a pregnant woman could take AZT without it negatively affecting her baby. Adults can take AZT because most of our cells are not synthesizing DNA, unlike those of bacteria (Singleton, 2012). But, a baby would have a much higher DNA synthesis rate than an adult, so couldn’t a fetus therefore be adversely affected? Since a mother’s blood doesn’t mix with her fetus’s during pregnancy does that mean her baby is safe from negative side effects? Because I am very interested in doing humanitarian work, especially in Africa where the AIDS rate is high, I decided to do some research on AIDS patients and AZT’s.
    The first website I discovered was a drug site that offered information about Zidovudine, a drug which contains AZT. Zidovudine is given to mothers with HIV so they do not give the virus to their child during the birthing process (Zidovudine, Retrovir, 2010). I also noticed that the drug information for Zidovudine included dosage amounts for infants and young children too. I thought this was really interesting, especially since a fetus and a small child would certainly have more developing DNA than an adult, and would therefore be assumed to have more negative side affects from the drug. Even if a fetus is protected by the fact that his or her mother’s blood does not mix while in utero, a young child would certainly be receiving the “full effects” of drug when taking it “out of utero.”
    Next, I decided to pursue a study done on the effects of Zidovudine in pregnancy published by the New England Journal of Medicine. In this study, it was concluded that while Zidovudine resulted in very adverse effects in mice embryos, overall it did not harm human fetuses (Sperling, et. al, 1992). Despite this result, it was also taken into consideration that many of the mothers and children treated with Zidovudine during pregnancy were HIV positive, and as a result had many other factors which could affect their baby-both before and after birth. Additionally, since Zidovudine has not been in use very long, many of its later effects may not be known (Sperling, et. al., 1992).

    After learning that in most cases Zidovudine did not negatively affect a fetus, I decided to investigate the logistics of why. Unfortunately, I could not find any “exact answer” to my question, but I did find some answers as to how exactly Zidovudine works which helped me to form a hypothesis myself about why it does not seem to harm fetuses.
    Similar to the AZT facts we discussed in class concerning bacteria, Zidovudine works to limit the ability DNA synthesis to occur (Monson and Schoenstadt, 2010). Just like bacteria, and any other living thing, viruses need DNA to survive. However, the HIV virus is different in the fact that it must first change its RNA to DNA to replicate. This process is done by using the protein reverse transcriptase enzyme. Zidovudine works in the body to “trick” reverse transcriptase into thinking it is a building block used to make DNA. When the reverse transcriptase attempts to use the Zidovudine to make DNA, it does not work (Monson and Schoenstadt, 2010). Based off of this information, I hypothesize that because the Zidovudine is “selectively toxic” to the HIV virus it does not (or has not been proven to) disrupt a human’s DNA synthesis enough to cause drastic side affects-in mother or child.


    Singleton, David. Microbiology Lecture. York College of Pennsylvania. 24 February 2012.

    Sperling, R. S., et al. “A Survey of Zidovudine Use in Pregnant Women with Human Immunodeficiency Virus.” The New England Journal of Medicine (1992): 857- 861. Web. 25 February 2012.

    Monson, K. PharmD and Arthur Schoenstadt MD. “Zidovudine.” 2010. eMedtv. Web. 26 Feb. 2012.

    “Zidovudine, Retrovir.” 2010. Web. 26 Feb. 2012.

    • An excellent spot to start research about medical research is the National Center for Biotechnology Information (NCBI), through the National Library of Medicine website at the NIH. I ran a quick search there in the Pubmed database, and found a fair number of hits on the cytotoxic effects of nucleoside analogs on fetal tissue. Here for instance is a review article from 2004, and the take home message from the review is that AZT and other anti-HIV therapeutics DO incorporate into fetal DNA. I think that there are several things going on, as Emily noted above. First, the placenta may prevent some of the drugs from going from the mother’s circulation over to the baby’s circulation. Second, metabolism of nucleoside analogs is going to be somewhat different in the developing baby than in the mother. But, as the review article indicates above, pharmacologically-relevant levels of these medications in primate animal models do result in measurable uptake into the developing fetus, and these levels do result in a number of detrimental cellular issues, including cell death, tumor development, and telomere shortening, which is involved in the overall lifespan of those cells.

      When Emily raised this very important point on Blackboard, I also commented there that one other statistic also comes into play here, which is also mentioned in the review article linked up above:

      Administration of zidovudine (AZT) prophylaxis during pregnancy reduces the rate of infant HIV-1 infection to approximately 7%

      and then you have to weigh the positive benefits to the baby (HIV infection drops to a remarkably low level), against the potential detriment to the baby. And that is some math I can easily get behind!

%d bloggers like this: