Friday, September 10, 2021

Reinventing Oligonucleotide Synthesis

 Wow, where to begin,

Our latest endeavors on the P(V) front are now published in Science. Now, before talking about the research, I must begin by giving a shout out upfront the (rather large) team who is responsible for everything in the paper. It was an incredible collaboration and everyone had a crucial role to play.


Bristol Myers Squibb: Yazhong Huang, Shenjie Qiu, Bin Zheng, Stephen Mercer, Richard Olson, Mike Schmidt, Ivar McDonald and Martin Eastgate

TSRI: Wei Hao, Julien Vantourout, Natalia Padial, Javier Lopez, Rohan Narayan, Donna Blackmond  and Phil Baran
After the team’s initial work developing the PSI reagents for the stereocontrolled synthesis of phosphorothioate linkages, much work remained to build out the P(V) platform, both to include other types of desired P-linkages but also to develop a protocol for using the novel reagents on an automated oligo synthesizer. The development of these additional reagents PS2, PO and RacPS all proceeded in different ways.

The challenges overcome in the syntheses of the PS2 and RacPS reagents were a simpler story so I'll direct you to finding that info in the publication. On the other hand, there is the P(O) reagent. To the best of our knowledge there had never been a published P(V) reagent capable of forming a phosphodiester bond between two nucleosides with phosphoramidite-like kinetics (i.e. crazy quick). The initial hit for such a reagent came in July of 2018, followed by my famous last words: “shouldn’t take long to find the perfect P(O) reagent” Although the hit was exciting, it required KOtBu as a base and wasn’t compatable with DBU or organic bases that would be amendable to an automated oligo synthesizer. 

To address this, we evaluated new P(O) reagents on several criteria:

1. Could they be made (Synthesis)

2. Could they be reacted with a nucleoside to afford a monomer (Loading)

3. Could they be coupled via an organic base with a second nucleoside to afford a dinucleotide  (Coupling)

We had two ways of making the reagents, either synthesizing 1,3-mercaptopropanols and reacting them with POCl3 or by generating analogous compounds with 1,3-diols and PSCl3 followed by an interesting atom transposition using TBAI. We found a lot of things that could work (and plenty that couldn’t) but only one scaffold was able to accomplish all 3 of our goals. P(O)-5 the diphenyl scaffold.  After this discovery we investigated the aryl electronics and stereochemistry of the diphenyl backbone. Attempts to modulate the coupling efficiency through the aryl electronics proved negligible. 

The stereochemistry was a more interesting questions, because the first synthesis of this reagent used a racemic mixture of the mercaptopropanol, generating a racemic mixture of P(O) reagents which upon loading onto a chiral nucleoside revealed a mixture of the possible isomers that one could imagine. We began synthesizing single isomer versions of these reagents and observed that the Syn scaffold epimerized into an inactive form during the loading step. 

Further investigation of the anti-backbone revealed that the (R,R) substituted system gave the fastest coupling times. We next looked into the synthesis of this reagent, the stereochemistry could be programed in via a Noyori reduction and using PFP as a leaving group provided suitable physical properties to afford a solid reagent. Despite this success, the reagent still had two flaws, it wasn’t stable enough (could only survive about a week on the benchtop) and it was still a mixture of isomers, with one reacting much faster than the other. 

Sitting around the lab one day, we thought to ourselves, why don’t start with the original PSI (a reagent we knew reacted at lighting speed) and swap the S for O? It seemed rather straightforward. In practice we were never able to do this successfully on the PSI reagent itself so we tried it on the PSI loaded monomers instead. Although we could succesfuly perform S-O transpositions on these monomers, there were always side reactions associated with the transformation, ultimately leading us to abandon this approach. 

Third times a charm as they always say, so revisiting the idea of performing the S-O transposition on the reagent itself, we knew we the reagent formed with PFTP (PSI) was too reactive after making the P(O) version of itself. So after generating PSI alternatives with different leaving groups, we found that we could successfully perform this transposition to generate a stable reagent, most notably the 4-bromothiphenol version.

There’s plenty more to the story. We evaluated the kinetics of the P(V) molecules, demonstrating that they are on par with phosphoramidite chemistry, we developed a novel universal linker capable of performing oligo synthesis under the unusual basic conditions found in the P(V) protocol and we even took a trip to Cambridge MA to visit our collaborators at BMS and see and learn first-hand how they were making oligonucleotides.

Also, if you’ve read this far it’s fair to assume you’re a fan of the P(V) approach to oligos so check out our perspective on P(V) chemistry that was also just published in ACS Central Science!


If you’re still reading, and want to follow the next steps of the P(V) story, keep an eye out on or drop one of us a line to check out how were using the P(V) technology (among other things) to develop the next generation of oligonucleotide therapeutics and “replace dogma with data”.  – Kyle and the P(V) Team

Thursday, April 15, 2021



Author information:

Chuanjun Jiao, IKA Works, Inc. 3550 General Atomics Court, MS G02/321, San Diego, CA 92121-1122, United States



Note: This report is not related to the recently added new function “Rapid Alternating Polarity (rAP)”. The change described here is relevant only when the default “alternating polarity” function is used together with a reference electrode. Currently, only “alternating polarity” function is supported to be used with a reference electrode, and rAP function is not compatible with a reference electrode.  In addition, further software update is not necessary to a unit in which the rAP function is already installed, since the latest software supporting the rAP function already includes this change.  



Recently the Reid Group at University of Strathclyde reported a mechanistic study of electrochemical benzylic oxidation with the use of ElectraSyn 2.0 (ChemElectroChem 2020, 7, 2771). During their study, the group observed unexpected product distribution change when alternating polarity was applied during a constant voltage experiment with ElectraSyn 2.0. The report describes their careful investigation of this phenomenon, and casts a practical consideration on users of IKA ElectraSyn 2.0.  

Here is the key points of their report:


·       Unexpected outcome was obtained under the condition of constant potential reaction mode­– reference electrode ON–alternating polarity ON.

·       During alternating polarity, the same magnitude of current would be expected in both positive and negative phase. However, the observed current magnitude was totally different between these two phases.

·       This asymmetric current distribution affected electrochemical reaction, resulting in unexpected product selectivity change.

·       Therefore, ElectraSyn 2.0 might have an engineering concern.   

Summary of IKA investigation


IKA conducted a series of experiments to clarify the nature of this phenomenon and to improve ElectraSyn 2.0. The detail of our internal investigation can be found in Detail of our investigation in this document. Here we would like to provide a summary of our investigation. 

·       We verified asymmetric current distribution in the same setting described in the report (constant voltage–reference electrode ON–alternating polarity ON).

·       This asymmetric current distribution was observed only when reference electrode is ON. In other words, this issue does not occur when reference electrode is OFF. 

·       This is simply originated from how voltage is controlled in ElectraSyn 2.0. 

·       Accordingly, ElectraSyn 2.0 does not have engineering concern.


Software update

Although we confirmed that ElecraSyn2.0 does not have any engineering flaw, we updated ElectraSyn 2.0 software to enable equal current distribution in the above experimental conditions (constant voltage–reference electrode ON–alternating polarity ON). 

When this setting change is needed


A user needs to enable this function only when Constant Voltage reaction mode with Reference Electrode ON and Alternating Polarity ON  


Effect on past research 


The issue only occurs when Constant Voltage reaction mode with Reference Electrode ON and Alternating Polarity ON. In other conditions, current and voltage control is accurate and experimental data is reliable.

Detail of our investigation 


IKA ElectraSyn 2.0 provides many functions that satisfy different requirements for electrosynthesis. Among those functions, Constant Voltage is a reaction mode where the reaction is executed at a fixed voltage. This reaction mode is generally more selective than Constant Current mode, though the reaction time tends to be longer. Alternating Polarity is an option that periodically switches anode and cathode. This option is useful when electrode fouling is observed. Switching polarity helps to alleviate accumulation of deposit. Both Constant Voltage and Alternating Polarity are routinely used for electrosynthesis. 


The recent report by the Reid group described a mechanistic study in synthetic organic electrochemical method development with the use of ElectraSyn 2.0 (ChemElectroChem 2020, 7, 2771). In this report, unexpected product selectivity change was observed in electrochemical benzylic C–H oxidation of toluene derivatives by employing Constant Voltage alongside Alternating Polarity function equipped in ElectraSyn 2.0.


During those experiments, it was observed that the magnitude of the current during the second half of alternating period was greater than during the first half; giving an asymmetric current over time response. This is not desired because alternating polarity is simply an electrode cleaning technique and should not alter the magnitude of current.

Further in-depth investigation of cell potential with alternating polarity revealed that the voltage during the negative phase was not controlled. The set voltage was 1.5V, which means the reaction voltage should alternate between 1.5V and -1.5V. However, the observed voltage was clearly not in this case.

In additional tests, a rapid degradation of NHPI was observed in Alternating Polarity experiments due to this unexpectedly high current input. Such degradation leads to the unexpected product selectivity.


Accordingly, the group published their findings to highlight an equipment engineering concern that is likely to influence and inform optimization strategies for a wide range of synthetic organic electrochemical methods under development.


To ensure product quality and to further improve EletraSyn2.0, IKA has conducted a thorough inspection of the instrument as well as detailed investigation of this issue. It turned out that the varied cell potential was caused by the way ElectraSyn 2.0 controls voltage against the reference electrode.


Method of voltage control


In the default setting of ElectraSyn 2.0, potential difference between Working Electrode (WE) and Reference Electrode (RE) is controlled when Reference Electrode is ON. Such setting remains unchanged during the experiments. In the ElectraSyn 2.0 apparatus, the working electrode (WE) is installed to the left side of the reaction cell and the counter electrode (CE) is installed to the right side. 

However, WE and CE do not necessarily indicate anode and cathode in electrolysis. WE can be both anode and cathode. For example, if 1.2 V is applied, WE will play the role of anode and the voltage should be set as +1.2 V between WE and RE.

On the other hand, if -1.2 V is applied, WE will play the role of cathode.

Alternating polarity


During alternating polarity, what is expected is the cycle of applying +1.2 V between WE and RE, and then +1.2V between CE and RE.

However, in the current ElectraSyn 2.0 software, the voltage control remains between WE and RE during the entire period. This means that applying -1.2V between WE and RE may induce completely different reaction, resulting in the observed asymmetric current behavior. 

In other words, if ElectraSyn 2.0 could switch WE and CE during alternating polarity, the problem of asymmetric current will be solved. This function was installed in an updated software.


Software update


After our investigation in software development, the aforementioned voltage control method was implemented. Users are provided with the option of enabling/ disabling the alternate polarity, with the latest software version (0.0.026 / 0.0.027).


The new software version was validated by using following test. In the first test, the solution containing 0.114 M Bu4NClO4 with acetone/MeCN = 1/1 ratio was used. As pointed out by the Reid group, constant potential mode with reference electrode-ON and alternating polarity-ON gave asymmetric current (blue circle in Figure 15.). When the setting is changed as illustrated above, totally symmetric current was observed (orange circle).

This test was conducted with another solution, verifying symmetric current behavior in this case as well. 

Accordingly, we have concluded that, alongside the existing function, an updated software version (0.0.026 / 0.0.027) will enable ElectraSyn 2.0 to perform a constant voltage reaction with reference electrode under alternating polarity conditions.


We thank Dr. Marc Reid, his colleagues, and all our users for supporting ElectraSyn 2.0 and providing valuable feedback.

Monday, October 5, 2020

Adventures in P(V) Chemistry

Our two recent P(V)-based projects that were just published in JACS and ACS Central Science were a lesson of teamwork and sometimes “serendipity” that we would like to describe in the following blogpost. 

Having worked on the P(V) reagent platform since its inception, I have spent a considerable amount of time sitting in front of the only NMR spectrometer at Scripps capable of detecting the 31P nucleus. Since these blog posts are supposed to be the behind the scenes version of papers we publish, I thought I would take this time to walk you through the fun journey from the Baran 

lab to the Molecular Biology building where one of our NMR labs is.  You begin by leaving the 4th floor of the Beckman Center for The Chemical Sciences, traversing the intricate Hogwarts style staircases down to the lobby. After exiting the building and making a daring leap over the crosswalk between the two buildings, it’s time to go further into the depths of the MBB building. Okay, back to the science…as you can see in the figure below, the loading and coupling events between the P(V) reagents and alcohol nucleophiles proceeds with a rather boring and predictable outcome via 31P NMR. Over the course of ~3 years, hundreds of compounds have undergone this reaction sequence. Regardless of the compound in question, the only observable peaks are 100 ppm for the loading and 55 ppm for the coupling products. The reactions between nitrogen and sulfur nucleophiles were never observed. 

This led us to the realization that the P(V) reagents could potentially solve the challenge associated with Serine selective functionalization as highlighted in the chart below.

At that time, Julien joined the P(V) team and first performed a series of competitive experiments between Serine and other nucleophilic amino acid residues. The selectivity observed was excellent and after a quick optimization campaign we applied the reaction conditions to the functionalization of linear and cyclic peptides. The chemistry proved to be really robust and afforded really good yields even on complex structure such as Vancomycin. Then, Prof. Bernardes came to Scripps to give a lecture and after a meeting with Phil, he was really excited about applying the new method to the functionalization of proteins. One of his students, Srinivasa, successfully functionalized ubiquitin and repressors 434 demonstrating the broad applicability of our Serine selective P(V)-platform. Other than its broad scope, one thing we really like about this new method is the ease of running and analyzing the reaction. A simple 31P NMR (1H coupled) will allow you to know if you functionalized the Serine residue over other nucleophilic amino acids by simply looking at the chemical shift and the multiplicity. The full details of the work are in the paper of which my favorite part is the computation work done by our amazing collaborators at BMS (Thanks again Antonio!!) which really explain this surprising level of chemoselectivity.

Another great application of the P(V)-platform has been its association to RASS (developed by the Dawson lab) to selectively functionalize DNA. It all started a few years ago when Dillon was working on DEL reactions. He would occasionally notice side reactions occurring at the terminal alcohol of the DNA head piece. This inspired him to dive deeper into developing selective reaction at that position. A year passed with many unsuccessful strategies and reactions explored… And just when hope seemed lost, he met PSI and learned about its exquisite reactivity. He decided to combine it to RASS and that’s how the SENDR platform for site selective DNA modification was brought to life (not without some hiccups). From there, the number of application idea flowing from the team grew with every day and many, many, many pilot experiments were pursued. Only a fraction of those made it into the paper, but as new ideas popped into our heads new collaborations were formed and exciting avenues explored! This project allowed us to dive into the powerful world DNA technology and has thus spurred exciting ongoing projects within our labs! These projects have allowed us to explore and participate regions of biomedical science that seemed completely foreign to us chemists before (we had never seen a DNA sequencer until a few days ago) and is enabling some remarkable science. We hope that this chemistry would enable the community to explore previously intractable and increasingly creative experimental designs! 


All in all, both of these papers were the result of many years of work across multiple labs (across the U.S and the Atlantic) and I am indebted greatly to the teams for not only the work but also the lessons and memories.


-Kyle and the P(V) team 

Monday, July 20, 2020

Tagetitoxin: Total Synthesis with Friends

Our recently completed synthesis of tagetitoxin was just published in JACS. I think it’s safe to say that this work is unlike anything we’ve done before in our lab. This is true not only from the standpoint of the chemical problems presented to us (see the paper, the SI, and some of the stories below), but also from a team standpoint, for everyone on this project is actually very good friends, both in and out of the lab. We therefore decided to use the space below for each friend on the team to write a little bit about their experience working with this beast of a natural product. Warning: long blog post ahead!

Hang Chu (Oct. 2016 – May 2018)
While searching for another molecule to pursue after the thapsigargins, Phil sent me Porter’s paper, which disclosed the most recently revised structure of tagetitoxin (tgt) at the time. Having primarily worked on terpenes at that point in my Ph. D., tgt seemed like a suitable molecule. As I recall, I wanted to work with something that had a nitrogen or two. Tgt certainly satisfied that requirement. As a bonus, throw in some oxygens, sulfur and phosphorous. Sure, why not? 

One of the toughest challenges initially was coming up with a reasonable retrosynthetic analysis for tagetitoxin. While there were not that many logical ways to take apart the molecule, finding an efficient and creative strategy to build the densely functionalized cyclopentane posed a challenge. In the SI you’ll find the evolution of our strategy as problems arose, so I will not belabor an extensive explanation of our logic. Our initial retrosynthetic analysis is shown below. Long story short, we foresaw three distinct challenges at the start of the program: 1. Core construction, 2. Cyclopentane functionalization and 3. Cysteine incorporation. 

 While there were many interesting snippets of problem solving in the first generation synthesis as disclosed in the SI, I’ll focus the two key reactions I had the pleasure to work on – 1) the oxidative furan rearrangement and 2) the thio-Claisen (see David’s section). 
To the best of our knowledge, there was no direct way in the literature to construct this type of amino hydroxy substituted cyclopentenone at the time (about 4 years ago..time flies). As the scheme illustrates, we had considered a couple of other ways to build this unusual cyclopentenone (double Claisen from a symmetrical aldehyde surrogate or an aziridine opening). These approaches were plagued with stereochemical issues and potentially harsh reaction conditions. We focused in on the β-hydroxy cyclopentenone portion of this building block and recalled that the Piancatelli rearrangement was an effective sequence to access these types of structure. Adding in the remaining functionalities took us back to a furanyl amino ester intermediate, which required the introduction of an equivalent of oxygen to get to the desired oxidation state. The problem-solving sequence to get this reaction to work initial was disclosed in the SI. Long story short, a non-basic reductant (dimethyl sulfide) was critical for this highly sensitive reaction to work. It has been a number of years since the discovery of this sequence so I’m a little cloudy on the details. I do recall isolating a wide range different retro-aldol initiated products along with a number of colorful decompositions. Also the use of the trifluoroethyl ester was essential to promote 1) good stability of the amino ester starting material and 2) the formation of the desired enone.

David Kossler (Mar. 2018 – Jan. 2019)
Since the synthesis of tagetitoxin was just published, I have the opportunity to reflect on my time as part of the team. After a short stint in the lab, working on method development, I joined Hang at the beginning of 2018. As we were office mates, I already had a sneak view on what was going on. By that time, he had found a neat way to build up the cyclopentane core via an oxidative furan rearrangement. As the stereoselective installation of the sulfur was problematic, he envisioned a relay via the allylic alcohol. It was clear from the beginning that this project was technically challenging in terms of sensitivity of the substrates, requiring careful adjustments of reaction conditions and workup procedures. Small deviations could lead to disastrous outcomes, and this happened. After we fixed a couple of issues in the first steps, the precursor for the [3,3]-rearrangement could be obtained in high purity. To our favor, once running the thio-CDI step in acetonitrile, the minor diastereomer from the Luche reduction reacted first with TCDI, and then cyclized intramolecularly, which made it easy to remove by column.

 Subsequently the rearrangement became a robust and clean transformation. This period was one of the most enjoyable, with both of us as a team trying to get as far as possible with the project. The dihydroxylation proved to be tricky as well, as the osmate cleavage led either to decomposition or protecting group swap without unveiling of the free alcohols. So we decided to install the sidearm first and then conduct the osmylation afterwards. All the details are in the comprehensive SI. Hang wrote his thesis alongside the lab work and graduated in mid-2018, upon which I continued with the project. Eventually, the free diol could be accessed, and the crystal structure confirmed the correct stereochemical alignment of all substituents. Big shout out at this point to the X-ray facility at UCSD, who perfectly resolved all structures I brought over during my postdoc, which was really helpful. 

A special moment was the first assembly of tagetitoxin’s skeleton. When analyzing the crude NMR, and in between all signal peaks, locating a set of multiplets that made sense for what we were trying to synthesize. Moreover, the comparison with the reported coupling constants for the natural product matched beautifully. With the trans-6,5 bicycle in hand, this was also the first proper evaluation point to see that the structure proposed by Aliev and co-workers was indeed the correct one. If the Js would have not correlated at this stage, it would have raised a big question mark that we were following the real structure of tagetitoxin.
 But in synthesis, for every problem you solve, you directly have the next step ahead of you, with the additional hurdle to bring material even further to the frontline. At this stage of the project limited material throughput became a real concern and hampered the advancement tremendously. I spent time on having a first look into the phosphorylation and the enantioselectivity challenge, but at some point the end of my Postdoc approached and I decided to go back from San Diego to Europe to finally enjoy some rain. Phil decided to hand the project over to Chi and Tom, who were about to finish the Herqulines. In order to not lose valuable knowledge and gained experimental details in this transfer, I assembled all available data and infos. Finally, in my last weeks we ran through the sequence together once. From then on, this well-oiled machine went on to make to target. Over the course of the last year I could still follow the updates and see the project evolving, one obstacle after another being removed. Looking on the finished sequence, it’s great to see how everything fell into place after extensive experimentation and many learnt lessons. I have nothing but huge respect for the final team in this relay run.

Tom Stratton (Jan. 2019 – Jun. 2020)
Fresh from our recent herquline victory lap, Chi and I were confident we would quickly shut the door on this project. We were dead wrong and spun our wheels for the better part of 18 months trying to understand what makes these molecules tick. I personally spent much of this time optimizing the key bromocyclization step, a reaction that will haunt me for the rest of my days. Since that story is (mostly) told in the SI, I will instead use this space to talk about how much I’ve learned from my good friends.

When I was a (highly) clueless first year grad student, Hang took notice. Right when things were starting to get truly ugly, Hang swooped in like a mama bird, brought me to his nest in BCC-439, and fed me tidbits of knowledge directly from his chemistry beak on a daily basis. He was a true mentor to me then and now, and a is also really strong dude.
I was assigned to be David Kossler’s (or Herr Doktor… show some respect) “temporary lab buddy” when he first arrived at Scripps from Europe. Lucky for us, the exact time component of “temporary” is ill-defined, as we still share cold beers together to this day, albeit from across the pond. The fact that we got to overlap (for one week) on this project was incredible as I got to see David’s remarkable precision, organization, attention to detail, and sense of team work first hand. We were lucky to have the wealth of knowledge gained by Hang and David at our disposable, mostly to the incredible work of my temporary buddy. Prost!
Kelly joined the project for a relatively short time, but in doing so, solved a problem that was a “non-starter” in terms of wrapping this project up. That is, stoichiometric osmium was required to effect dihydroxylation, and Kelly figured out the solution in literally a couple of weeks. This is no surprise. When Kelly was interviewing at Scripps, we recruited her hard. Not only did she possess several years of experience as a process chemist, she displayed the poise, focus, and kindness that is required to be a successful student and excellent teammate. Thus when my new hoodmate Kelly was recruited onto this project, Chi and I were absolutely elated and she stepped up to the plate big time. 
Dillon M. Flood has saved my ass more than once. Whether it be stuck on a 50sketchy left traverse at Big Sky, or at the outer reef of Teresa’s when my leash snapped and the waves were the size of a small cathedral, Dillon has been there to bail me out. Thus, it should be of no surprise that he rose to the occasion and contributed so much to this project in its final stages. We had intended to use a CRO to run the RNA polymerase assay, and when I asked Dillon for recommendations, he said “dude we just got a new plate reader and I can do this for you next week.” Having already showed Chi and I the ins and outs of anion exchange chromatography, he decided to bring his impact on this project to the next level. Sure enough, one week later he was sending us beautiful data depicting that (+)-tagetitoxin is the sole enantiomer active against E. coli RNAP. What a guy! I’ll see you up north, my friend.
There are simply no words that can describe how I feel about my partner Chi, but I’ll do my best. I doubt there is a more patient, humble, brilliant scientist out there than my friend Chi. Thank you for everything you have given me. I hope I have been able to return even a fraction of it. 
Finally, Phil is the best mentor I could have ever asked for. I truly feel like I won the PI lottery. Phil has a passion for exploring the unknown that is contagious and these ideas transformed the way I have thought about solving problems in chemistry forever. 

Kelly Eberle (Sep. 2019 – Jan. 2020)
I certainly feel lucky to have started in the lab at a time when there was still work to be done on Tagetitoxin. I knew that I wanted to do total synthesis, was interested in alkaloids, and that ideally I would work on an existing project to ‘get my feet wet’. When Phil suggested I join this team it seemed like a perfect opportunity. As a first year who had worked in pharma before coming to grad school, I thought I would know what I was doing. I can assure you that feeling went away very quickly. Thankfully I had joined a team of very intelligent chemists that were kind enough to mentor me and show me the ropes. Tom’s down-to-earth realness and Chi’s endless positivity helped get me through some of my darkest and most frustrating days as a first year. Working on this project also exposed me to several reactions I hadn’t run or even heard of before coming to grad school. Spending my first five months on this team taught me so much about technique, time-management, and chemistry in general; I’m very fortunate to have this foundation to build off of for the rest of my time in the lab.
Some key takeaways from working on TGT: Always be bringing material forward while you’re working at the front line. Compared to my past life of working in a process lab, scale up in academia is tough and the glassware is heavy. Ask your teammates questions (especially when things aren’t working) because there’s always more to learn.

Chi He (Jan. 2019 – Jun. 2020)
After finishing the herquline project, Tom and I decided to keep working on total synthesis. At that moment, our friend Herr Doktor (David) was going back to Germany to start his professional career. Tom and I took over the tagetitoxin project from Herr Doktor, not only because of its intriguing structure, but also our friendship with Hang and David. 

Hang and David had worked on this project for over 1 year. Moreover, David had confirmed the desired stereochemistry of bromocyclization product by X-ray in the last week before he left (see above). To be honest, this good news made me and Tom very excited to think we could put a bullet to this project soon. In two short weeks, however, we would come to understand just how many more problems awaited us. In general, we modified and optimized the whole route in the past 1.5 years (for more experimental details and logics, please see SI). Here, I would like to share my experiences and feelings from 2 late-stage procedures: phosphorylation and the final purification. 
 Des-P tagetitoxin was easily generated by convenient workup after exhaustive hydydrolysis by Ba(OH)2. We were excited to find the chemical shifts and J-values in 1H NMR of des-P tagetitoxin were significantly similar or identical to the original data in the isolation paper, which indicated the structure we were working on was promising (non-trivial given the structure of tagetitoxin had yet to be confirmed). It seemed we could put a bullet in this project  in just one more step! However, the final selective phosphorylation of C8 alcohol failed after several attempts with chemical conditions. The failure was probably due to the poor solubility of des-P tagetitoxin in organic solvents. Although it might be possible to achieve the phosphorylation with the help of enzymes, we decided to pursue this final step using chemical methods. 
 To improve the solubility in organic solvents, we protected the free amino alcohol as the hemiaminal. After methanolysis, 18’ was obtained as an ideal substrate for phosphorylation, with two masked methyl esters and hemiaminal to improve the solubility and stability. To our surprise, a lot of conventional phosphorylation conditions did not work on this substrate due to either poor chemoselectivity or low reactivity. Fortunately, our lab had already developed a practical method to achieve the chiral phosphorylation by a talented graduate student Kyle. With the help of Kyle, we successfully put on P(V) with highly reactive (+)-Ψ reagent in excellent yield. The reaction condition is mild and highly reproducible. I want to highlight that the advantages of this method were not limited in phosphate installation, but also helpful in crystallization and chiral resolution. We confirmed the structure and absolute configuration through the X-ray of 19a and obtained both (+)-tagetitoxin and (–)-tagetitoxin in two more steps.
 I still remember the first time I got tagetitoxin. I saw a huge rainbow outside of window in early morning of 2019 Thanksgiving. After taking a 1H-NMR of this crude mixture, I was excited to tell Yuzuru I might have made tagetitoxin! One day later, I emailed Phil and my teammates about this exciting news after confirming by 13C-NMR. That moment, I thought we could finally put an end to this project and enjoy the Christmas holidays. However, it was just a beginning of the most challenging problem I experienced during this project: the final purification. 

The natural product contains so many polar functional groups on a tiny skeleton, which makes it much more polar than peptides or even palau’amine. Attempts to purify with preparative HPLC were proved intractable as the highly-polar mixture eluted at very beginning. Luckily, our friend Dillon, another talented graduate student, gave me a hand at that moment. After trying several different purification methods, we found that anion exchange chromatography with DEAE resin was optimal. However, the only way to confirm the purity was by taking NMR of every fraction after lyopholization (LC/MS, for example, was not sufficient to determine purity). It was common that a single purification procedure could take several days, only to realize at the very end of the sequence that we had again failed. I forgot how many anion exchange columns I had tried, definitely >15. In most cases, the impurities came from the product and they were co-polar. Further, it was impossible to attempt re-purification from this mixture. The only solution was to carefully scale up a new batch of tagetitoxin and try a new purification. After several failures, we realized tagetitoxin is extremely sensitive to acid, as decomposition was observed at pH 4 for 30 min. After fully understanding this nuance, and combined with lots of effort and advice from Dillon, we finally collected clean NMR spectra of tagetitoxin! All told, the final purification took us almost half a year, finally enabling completion of this project. It was very challenging and at times frustrating for me to spend such a long time to deeply understand how fragile and tricky this molecule is. On the other hand, I am also very proud of our courage to overcome the array of challenges that were presented to us!

Finally, the most impressive thing for me in this project is the great cooperation between friends. I would like to thank the support and help from Hang, David, Tom, Kelly, Dillon and Kyle. The project would not be done without any one of you!! 🍻

Dillon Flood (Jan. 2020 – Jun. 2020)
Over the past few years, Tom and I have had countless conversations which generally revolved around south pacific storm cycles, where we were chasing swell that weekend, and who would pick up the Tecate before we left. Although never at the forefront, chemistry would bubble up through these interactions. The world of total synthesis inhabited by Tom and Chi felt foreign to me, so the usual griping about failed experiments, unrealistic deadlines, and long hours always seemed to need a bit of translation.

I’m not sure what initially provoked Tom and Chi to ask for my help with an anion exchange purification but it very well could’ve been during a collegial discussion on right way to drink one’s preferred Mexican beer. However it happened, Tom and Chi were showed up on the first floor of the Beckmann, bright eyed and bushy tailed, taking copious notes, and very ready to purify their compound that I didn’t want to ask about how long it took them to make… Although this should’ve been a routine method, it never is. This led to Chi spending better parts of some days in the Dawson lab perfecting his anion exchange chromatography, while always expounding on the genius of LeBron James, just so I was sure. 
At some point I made the mistake to ask Tom (while surfing at Blacks), what this compound even does? After a long story about phytotoxins, RNA polymerase, optical rotations, isolation chemists without compound, and a half-baked idea to spray plants (?!), all I could think at the time was that I needed another wave. But out of this conversation, and a little googling, came the idea of performing the in vitro assay on both enantiomers. At first, the pair was not so enthused when I mentioned the assay. But when I mentioned that I could run it for them, everyone was on board. And then again, Tom and Chi were back down on the first floor of Beckmann, compounds in hand, ready for me to tell them which enantiomer was active. At this point I had to ask them to leave, unless they wanted me to watch me mess this thing up. So after loading the plate to the Hamilton soundtrack, running the assay and reading the results, I was shocked the experiment had worked so well. One compound was active and the other was dead. But this was the beauty of working at Scripps, friends in totally different fields could halfheartedly bounce ideas off one another that may help resolve and unanswered question.