A tale of serendipitous discovery

ISRIB, the first cognitive enhancer

An unlikely start

Scientific advancement is unpredictable. Despite our best efforts, we cannot foresee where the next big breakthrough is going to come from. One thing has proven to be true: If we want to encourage discovery, we have to disseminate opportunities widely. While big collaborative initiatives like CERN, LIGO, the Human Brain Project or the War on Cancer are doubtlessly worthy efforts, they have to be an addition to, and not in place of, basic research.

Basic research is a phrase that is often thrown around aimlessly. Here, we refer to basic research as curiosity driven scientific inquiry, with no immediate payoff or application visible except for the gain in knowledge. Funding basic research was what enabled the recent CRISPR revolution, a fantastic deed with grandiose potential, but also another sad reminder of which important areas of foundational scientific inquiry might not find funding in the future anymore. Despite the numerous successes, we remain ignorant and wondering of how many breakthroughs we already denied in the ever decreasing support of basic science.

Why is basic research in decline?

Supporting basic research entails more than just putting up the money, it demands recognition of what Peter Walter calls “Walking Along the Serendipitous Path of Discovery”, allowing people to pursue their ideas and follow their passion.

Serendipity thus defines a process of “discovering something by accident and cleverness while investigating something quite different,” often leading to unexpected insights. -Peter Walter, MBoC, 2010

While general funding is often bleak, real pain is inflicted onto basic researchers by fiscal inequalities in what research does get funded, journal gatekeeping of what is deemed a sexy story or valid field of inquiry, and last but not least career scientists who usurp their funding niche and suppress newcomers with fresh and controversial ideas.

All these issues slowly but steadily corrode and continue to transform the scientific endeavor into an industry with its own rules, rather than celebrating the messy and deeply human process of discovery and exploration for its own sake. For knowledge. For curiosity.

Yet not all is lost. There are still many scientists out there who deeply understand and follow the serendipitous path of discovery and deserve our support and goodwill. It is our duty to empower them and tell their stories.

Especially since following that serendipitous path requires professors to take a personal, professional or financial risk, for example in putting their trust in a person who had been out of academia for almost a decade and with a gaping hole in her publication record. A deadly sin in today’s competitive ‘publish-or-perish’ world.

A complete breach of academic industry standards. An insurmountable shortcoming for many castaway scientists.

Still, despite all of this, professor Peter Walter decided to give it a shot:

Our discoveries are not linear, so why do we expect our careers to be?

With that in mind, he rehired a former PhD graduate going by the name of Carmela Sidrauski.

This is the story of her discovery.


Carmela started her scientific career with as an undergraduate research fellow in Buenos Aires, Argentina before moving to the US to perform graduate work at Tyler Jacks lab at MIT. After 2 years, she moved to California to join Peter Walters lab at UCSF as a PhD candidate investigating the unfolded protein response (UPR).

In short, the UPR is a quality-control mechanism that ensures only correctly shaped proteins move on to eventually perform their function in the cell. Proteins are complex arrangements of hundreds of amino acids in 3D-space, whose very shape allows them to perform the myriad of functions that enable life. It is their very folding which imbues proteins with unique capabilities; without it, proteins are just boring, dysfunctional strings of amino acids.

The UPR is triggered by the detection of unfolded peptides, which are sensed by three ER-localized transmembrane sensors, IRE1, ATF6, and PERK. Back then, in the pre-human genome times of molecular biology, the first signs of Carmela’s creative brilliance significantly contributed to our understanding of IRE1 function. In fact, she had discovered one of the freakiest mechanism of cell biology, unconventional splicing of the transcription factor HAC1. Unconventional because it occurs without the spliceosome, a real catch of something so unusual that has never been reported before.

Once IRE1 ( →HAC1) and ATF6 TFs are activated, they act to alter which mRNAs are being transcribed while PERK phosphorylation reaches the eukaryotic translation initiation factor 2 alpha (eIF2α) protein, a master regulator of translation. The field was in a hot streak, with many labs working on the process of how the cell senses and adapts to misfolded proteins.

Next, the serendipity of life interfered. Despite a successful PhD in an emerging scientific field, Carmela decided to leave research and start a family, for a time period which would finally amass towards more than eight years. After all these years, any chance of an academic career for Carmela had pretty much gone too.

In the meantime, our understanding of the UPR grew significantly, as did our knowledge in how the cell handles various stress conditions. Surprisingly, many researchers discovered that handling various cellular stresses depended on the phosphorylation status of the eIF2α protein, a small but critical component of the translation pre-initiation complex (PIC), which has the power to shut down global protein synthesis, while enhancing translation of a small subset of mRNAs. In fact, the ability of eIF2α phosphorylation to shut down translation is so powerful that multiple cellular-emergency signal cascades lead directly there.

eIF2α phosphorylation is a critical event for a cell when implementing the integrated stress response

Thus UPR research has found it’s way into a bigger phenomenon, the socalled “integrated stress response”, with eIF2α phosphorylation status serving cells as decision checkpoint on whether to adapt cellular behavior or commit suicide to protect the rest of the body.

In metazoa, distinct stress conditions activate different eIF2α kinases (PERK, PKR, GCN2, and HRI) that converge on phosphorylating a unique serine in eIF2α. This collection of signaling pathways is termed the ‘integrated stress response’ (ISR). -Sidrauski, Elife, 2013
It was amidst these new findings when Carmela decided to get back into the game, so she contacted her old professor.

A fateful conversation later Carmela was re-entering the field; again in Peter Walter’s lab.

With her background in UPR, she decided to investigate the basic mechanisms of PERK signaling further. She performed a cell-based screen to identify inhibitors of PERK signaling, a multistep process starting with 106,281 compounds covering a wide chemical space looking to induce a stress-responsive luminescent reporter.

Big chemical screens are the equivalent of finding the needle in the haystack. Figure 1, Sidrauski et al., 2013

The initial screen yielded 460 potential hits, which needed to be further validated for PERK specificity, cell toxicity and a host of other pharmacokinetic parameters. From a final set of 28, none of the compounds seemed to be viable for usage in vivo. In fact, this is very common for large-scale pharmacological screenings, that many promising lead compounds turn out to be unfit in the end; which is part of the reason why drug discovery & development is such an expensive endeavor only multi-billion pharma companies can risk. A uniquely frustrating experience.

Yet serendipity acts in mysterious ways.

According to a recent talk given by Peter Walter in Zurich, Carmela haphazardly went back to the “trash pin” of her compounds and picked out ISRIB, one of the last 28 candidate chemicals. The reason ISRIB was initially discarded was because of its poor solubility, making it impossible to administer sufficient amounts to animals/patients. At least this is what common wisdom dictates.

However, it turned out that ISRIB has an extremely high cellular potency [ trans-ISRIB’s IC50 is about 5 nM], which meant that at equal concentrations, ISRIB was already a 1000x more effective than most other primary compounds. Accordingly, one could use a thousand times less of it and still get the desired effect. An easier way of phrasing it: For ISRIB, solubility was not a going to be problem.

ISRIB is a potent inhibitor of the integrated stress response. Figure 2, Sidrauski et al., 2013

So much for not giving up easily.

Having discovered ISRIB, Carmela was due to some surprises once she started further characterizing the compound. First off, she realized that ISRIB impairs the PERK signaling pathway but has no effect on PERK itself or PERK phosphorylation. In effect, PERK phosphorylation seemed to be even increased.

Next, she assessed whether ISRIB blocks the phosphorylation of eIF2α, this master-regulator of the pre-initiation complex.

Unexpectedly, ISRIB did not block eIF2α phosphorylation under either of these ER stress-inducing conditions. […] These results indicate that ISRIB blocks effects downstream of PERK and eIF2α phosphorylation. -Sidrauski, Elife, 2013

Carmela and everybody else in her lab was stunned. They have shown that ATF4 production is inhibited by ISRIB, yet everything upstream, including eIF2α phosphorylation, is fully intact.

It made no sense. Something was missing.

Further experiments showed that protein translation was still intact upon ISRIB treatment, which goes against all evidence of eIF2α phosphorylation being the critical marker for translational shut-down. They speculated that somehow, ISRIB rendered the cells insensitive to the phosphorylation event, masking it. In support of that theory, they also tried if ISRIB could counteract eIF2α-independent translation inhibitors (mTOR-S6K1), which it could not.

Additionally, they reasoned that if eIF2α phosphorylation is the critical event, it should not matter if they induce PERK or other kinases like PKR, GCN2, and HRI of the integrated stress response. It did not matter.

They concluded:

ISRIB counteracts all cellular stress pathways that utilize eIF2α phosphorylation.

For the moment, they could not figure out how exactly ISRIB did what it did, in fact in would take another two years for them to unearth and publish the mechanism of action. [Sidrauski C. et al., Elife, 2015]

Once again, serendipity had other plans.

Shortly after ISRIBs inital discovery, Peter Walters received a call from colleagues of the McGill cancer center in Montreal, Canada. They have been working with eIF2α-[S51A] mutant mice, who are unable to phosphorylate eIF2α and had a curious phenotype: They displayed enhanced memory.

They were interested in getting a hold of ISRIB, since pharmacologial inhibition of eIF2α should in theory replicate this mouse phenotype. Yet Peter Walter’s lab had never worked in neuroscience and had no expertise in it, something that would make most career-scientists very wary of persuing neuroscience projects. Still, they had the guts to follow through. A bona-fide collaboration was born.

After pharmacokinetic profiling, it became clear that ISRIB would reach sufficient bioavailability for in vivo studies, even more importantly, it could cross the blood-brain barrier. They discovered that even a single injection of ISRIB was sufficient to deliver ISRIB to the brain several folds higher than its required IC50.

The real surprise came once they started assessing cognitive abilities in ISRIB treated mice.
ISRIB enhances spatial and fear-associated learning in rodents. Figure 6, Sirauski et al.,2013

So how does one test something complex like memory?

Assessing cognitive tasks is a hard and limited scientific inquiry. There are certain experimental setups that can test for very specific abilities, for example the ‘Morris water maze’-experiment, where animals learn to associate visual cues with the location of a submerged hidden platform.

As seen above in figure 6C, initially mice take a long time (>100 seconds) to find the hidden platform, but if you repeat the experiment over days, they start remembering where the platform is hidden and find it rather quickly. After only 3–4 days, mice treated with ISRIB started pulling ahead of untreated mice, finding/remembering the platform four times faster (escape latency after 5 days of training = 16.4 ± 4.8 s) compared to vehicle treated controls (68.1 ± 20 s, p<0.05).

Another way of assessing memory is via so-called fear conditioning. Here, the researchers paired a specific environmental context (a certain cage) with a foot shock (induces fear) and measured signs of fear (freezing of animals). Again, ISRIB treated mice showed increased freezing ( → memory of foot shock) of the contextual environment after 24h compared to the control mice. Lastly, they used an auditory signal combined with a foot shock and observed freezing. Again, ISRIB injected mice froze longer than controls once they remembered the trigger 24h later.

No matter what behavioral memory experiments the researchers tried, ISRIB consistently let the mice overperform.

Altogether, the researchers concluded:

ISRIB increases memory consolidation, allowing pharmacological enhancement of the brain’s ability to learn. Evolution therefore did not arrive at a maximally optimized process, imposing a brake (via eIF2α phosphorylation) on memory consolidation. This mechanism may underscore the importance of filtering memories before committing them to long-term storage. […]
Our findings raise the possibility that ISRIB or compounds with related activities could serve as invaluable tools in deciphering these higher order brain functions and perhaps be beneficial as a therapeutic agent effecting memory improvement in diseases associated with memory impairment. -Sidrauski et al., Elife, 2013
Finally, after a turbulent strive, serendipity, luck, tenacity and curiosity through years of basic research, Carmela and her coworkers ended up having discovered ISRIB, the first proven cognitive enhancer with a bright future.

Early fruits

Carmela’s breakthrough paper in 2013 has laid the groundwork for a whole array of recently published or soon-to-be published studies going deep into the implications and potential of ISRIB.

For example, a recent work led by Austin Chou et al. from the Brain and Spinal injury Center at USCF could show that ISRIB can be used effectively to treat traumatic brain injury (TBI) in mice.

ISRIB reverses memory failure from traumatic brain injury. Video credits to UC San Francisco.

Another study by Kabir et al., from Weil Cornell Medicine in New York, reported that ISRIB administration was able to reverse the anxiety/sociability deficit phenotype of their neuropsychiatric mouse model.

On the more basic research side, Stern et al., reported in vitro that ISRIB could partially mitigate neurotoxic damage from Elvitegravir, a side effect of the antiretroviral drug used to treat HIV patients.

Since memory consolidation is a critical process in many of our cognitive tasks, it is as of yet unclear how big ISRIB’s impact will turn out to be.

Obviously, there are also looming dangers; because ISRIB increases memory consolidation, it can have negative effects when combined with traumatic experience (PTSD) or addictive drug use. Something Peter Walter is also very concerned about. For example, he has shown that eIF2α-mediated translational control is critical in regulating cocaine-induced LTP (a hallmark of addiction!). Additionally, the thought of cognitive enhancers in the hands of overly ambitious parents will need to find a place for discussion. Lastly, and even more concerningly, some humans are already offering and selling the experimental drug in shady longevity forums without prior data of human clinical trials or on detrimental side effects and ethical concerns.

As always, with great drug potential comes great drug responsibility.

In summary, one element appears to be virtually certain:

We will come to hear more of ISRIB in the future. What it can and (also importantly!) cannot do. How many patient lives it will be able to improve. How much more knowledge, about biology and ourselves, it will help to generate.

For me as a budding scientist, I also hope we will remember that it started with one audacious professor’s willingness to re-hire an academic in exile and embark with her into curiosity driven basic research. This kind of courage ultimately exemplifies where scientific advancement really comes from: Our choices.

This story was inspired by a fantastic talk of biochemist Peter Walter, recent winner of the 2018 Breakthough prize in Life Sciences. His brilliance was only matched by his humility, as he was reserving all credit and appraisal for his students and coworkers, never for himself.



This story is part of advances in biological sciences, a science communication platform that aims to explain ground-breaking science in the field of biology, medicine, biotechnology, neuroscience and genetics to literally everyone. Scientific understanding has too many barriers, let’s break them down!


Do you have any questions or inquiries? Why not ask a scientist?