Download PDF
Synlett 2024; 35(19): 2155-2173
DOI: 10.1055/a-2222-1667
account
Isotopic Labeling

Method Development and Syntheses Examples of Isotopically Labeled Compounds to Foster Operational Excellence in Pharma Industry

Anna Sib
,
Volker Derdau  ∗
› Further Information
 
 PDF Download Permissions and Reprints All articles of this category


The manuscript is dedicated to Prof. Laschat at the occasion of her 60th birthday – thanks Sabine.

Abstract

The different topics and synthetic approaches in an isotope chemistry laboratory of a pharma company are described. Besides the challenges in the synthesis of long-lived isotopes such as 3H or 14C, short-lived isotopes such as 68Ga and stable isotopes such as 15N, 13C or 2H approaches for the isotopic labeling are also demonstrated. Furthermore, method development with emphasis on collaborations with academic groups to tackle the future challenges are discussed.

1 Introduction

2 Isotopic Labeling with Hydrogen Isotopes Deuterium (2H, D) and Tritium (3H, T)

2.1 Deuterium Labeling for MS Standards

2.1.1 Labeled Nitrosamines – The Hunt to Quantify Hazardous Impurities

2.1.2 Deuterated Drugs, an Approach To Improve Existing Drugs or To Find Opportunities in Drug Discovery

2.2 Tritium-Labeling Methods – The Fast Approach to Radioactively Labeled Compounds

2.2.1 Hydrogen Isotope Exchange by Iridium Catalysis

2.2.2 Ruthenium-Catalyzed HIE

2.2.3 Nanoparticles as Catalysts in HIE

2.2.4 Photoredox-Catalyzed HIE

2.2.5 HIE via Classical Radical Mechanism

2.2.6 Beyond HIE – Halogen–Tritium Exchange

3 Challenges in 14C-Synthesis Projects

4 Short-Lived Isotopes – The Need for Speed

5 Beyond Isotope Science – Late-Stage Functionalization

5.1 Examples of Late-Stage Functionalization for Peptides

5.2 Examples of Catalyst-Controlled Late-Stage Functionalization

6 Conclusion


#

Key words

isotope science - drug discovery - isotope exchange - CH functionalization - catalysis

Biographical Sketches

Zoom Image

Volker Derdau studied chemistry in Münster and Braunschweig (Germany) and obtained his PhD in 1999 with Prof. Sabine Laschat. He was a Deutscher Akademischer Austauschdienst (DAAD) funded postdoc in Prof. Victor Snieckus’s group (Kingston, Canada) before he started at Aventis Pharma Germany as laboratory head in Chemical Development. Today he is Department Head and Senior Distinguished Scientist in the Integrated Drug Discovery platform (Sanofi, Frankfurt) responsible for Isotope Chemistry, Center of Excellence. He is Editor-in-Chief of the Journal of Labelled Compounds and Radiopharmaceuticals (Wiley, IF 1.9 in 2023) and author of more than 70 publications, two book chapters, and owner of eight patents. He has co-organized some International Isotope Society (IIS)- European Division (ED) workshops and the global IIS-conference in Heidelberg 2012, and he is member of the IIS Board of Trustees (BoT) since 2012. In 2017, he was elected president of the IIS-European Division chapter and became global IIS president in 2021 and 2023. Furthermore, he was lecturer at the University of Applied Sciences in Darmstadt (2012–2018), the Provadis Academy Frankfurt/Germany (2021 to present) and is a scientific consultant for CROs. He was the recipient of the IIS-ED prize in 2007 and of the European Isotope Science Award in 2022.

Zoom Image

Anna Sib studied chemistry at the University of Bonn (Germany) and finished her master’s degree in 2015. She obtained her PhD in 2019 from Technical University of Munich (TUM), where she worked in the group of Prof. T. A. M. Gulder. From 2019 to 2020 she worked as a postdoctoral researcher in the Institut Català d’Investigació Química (ICIQ in Tarragona) with Kilian Muniz, where she was funded by the PROBIST fellowship of the EU. In 2020, she moved to Cambridge (UK) where she worked at the University of Cambridge as Marie-Curie sponsored postdoc in the group of Prof. G. Bernardes. After two years, in 2022 Anna started her career at Sanofi as Laboratory Head in the Isotope Chemistry Department as part of the Integrated Drug Discovery team (IDD).

1

Introduction

Isotopes from carbon or hydrogen are unique tools for identifying and understanding biological or chemical processes. These isotope labeling allow for a direct incorporation of additional mass units or radioactive tags into an organic molecule with almost no change in its chemical structure, physical properties, or biological activity. Using deuterium (2H, D), carbon-13 (13C), or nitrogen-15 (15N) labeled isotopologues to study the unique mass spectrometric (MS) pattern generated from mixtures of biological relevant molecules drastically simplifies analysis and quantification in complex matrices. Such methods are now providing unprecedented levels of insights in a wide and continuously growing range of applications in the life sciences and beyond. Tritium (3H, T) or carbon-14 (14C) have seen an increased utilization, especially in pharmaceutical drug discovery or crop science industry. The efforts and costs required for the synthesis of labeled compounds are more than compensated by the enhanced molecular sensitivity for analysis and high reliability of the data obtained.[1] [2]

In this Account we discuss our achievements to further advance isotope sciences within the chemical society. Our main interests are covering methods for hydrogen isotope introduction (deuterium and tritium), late-stage functionalization with carbon isotopes (13C, 14C), synthesis of short-lived isotopes to develop methods for (68Ga, 11C, 18F), and to provide conjugates and tool compounds to study biological processes. Furthermore, radio- and medicinal chemists can combine their knowledge to develop useful procedures for late-stage functionalization of complex molecules to improve lead optimization but also to find new options for isotope labeling applications (Figure [1]).


# 2

Isotopic Labeling with Hydrogen Isotopes Deuterium (2H, D) and Tritium (3H, T)

Hydrogen isotope labeling of organic compounds is a well-known and important method in numerous areas, with special emphasis for drug development and crop science. The inherent properties of the stable and radioisotopes lead to different application. Deuterium on the one hand has a twofold increase in molecular weight, which induces a lower vibrational frequency in the C–D bonds compared to C–H bonds. This phenomenon leads to a higher activation energy for C–D bond cleavage in relation to its corresponding C–H bond and is known by the term kinetic isotope effect (KIE).[1] Using these physicochemical properties in drug design can help to improve the profile of a drug candidate. Deuterium incorporation decreases lipophilicity and increases metabolic stability at the labeled position. Thus, a replacement of hydrogen by deuterium may lead to an improved absorption, distribution, metabolism, and excretion (ADME) profile.[1] In addition, deuterated lead molecules are used as standards in bioanalytical mass spectrometry. Tritium on the other hand is the gold standard method for radioisotope labeling in early discovery stage. As hydrogen isotope exchange (HIE) is a much faster and cheaper method than 14C labeling, it is mostly used for in vitro studies and binding assays, and thus is essential for the early evaluation of precandidate drugs.

2.1

Deuterium Labeling for MS Standards

When performing bioanalysis with mass spectrometry detection an appropriate internal standard is essential for reliable results. The best option here is a stable isotopically labeled version of the parent compound that you want to quantify.1 When designing stably isotopically labeled substances, one is faced with the choice between different isotopes (e.g., 2H, 13C, 15N). Deuterium is the cheapest and synthetically simplest option for stable isotope labeling and is therefore the preferred choice. The deuterated standard should co-elute in liquid chromatography (LC) with the quantified compound and show a signal outside of the natural mass distribution of the analyte.[3] As the deuterated compound has a unique MS pattern, it is easy to track and will make the analysis more robust and reliable. For a given analyte of known molecular formula and mass, the molecular mass difference between the analyte and the stable-isotope-labeled (SIL) compound is determined by the number of isotopes incorporated in the structure. This number has to be high enough to minimize crosstalk between the analyte and SIL compound to a value less than 0.05%. A high number in isotope labels and a good chemical and isotopic purity of about 98% are necessary for a SIL standard. It is therefore preferable to have a sharp M + n MS signal rather than a broad spectrum. The synthesis of deuterated standards is often challenging, as building blocks are expensive or not commercially available at all. Thus, it is important to work on new hydrogen isotope exchange (HIE) or late-stage deuteration methods.

Zoom Image
Figure 1 Areas of interest in an isotope chemistry pharma industry group by example of Sanofi

Especially in the recent years progress has been made in this area. To facilitate the development of deuterium-labeling techniques, we teamed up with different academic and industrial groups. In collaboration with the Beller group, we developed a novel protocol for the catalytic hydrogen–deuterium exchange of biologically active tertiary amines.[4] By use of the so-called borrowing hydrogen methodology,[5] α,β-H–D exchange of a broad variety of amines can be achieved (Scheme [1]). The reaction is carried out using commercially available Shvo catalyst, a deuterated solvent as the deuterium source and either thermal energy or microwave technology. The right choice of deuterium source opens a broad spectrum of substrates, leading to high deuterium incorporation into the target compound and very high regioselectivity. The wide range of available deuterated solvent leads to an easy optimization process for different starting materials. We realized that isopropanol-d 8 is the best choice for most starting materials, especially for compounds containing primary and secondary OH groups, as it prevents oxidation to the respective ketones. However, the use of nonreductive deuteration reagents, like tert-butanol-d 1, resulted in better outcomes for substrates bearing functional groups prone to hydrogenation, e.g., double bonds or ketones. The robustness of this procedure was demonstrated by the deuteration of complex molecules and ‘real’ pharmaceuticals. It is therefore noteworthy that this method is a great tool to perform a highly selective H–D exchange of α- and β-N-alkyl protons. This is especially of interest as these positions are often sensitive to metabolic attack for many biologically active compounds. The method is therefore interesting not only for a stable-isotope-labeled analytical standard, but also for the development of new drug entities, using the kinetic isotope effect to enhance the pharmacokinetic properties.

Zoom Image
Scheme 1 Mechanism of α,β-H–D exchange in alkylamines

Next to the α- and β-alkylic positions in amines, we also developed a procedure for deuteration of heterocyclic amines like piperidines, piperazines, and tetrahydroquinolines.[6] This methodology bases on reductive deuteration of aromatic or unsaturated amines with deuterated ammonium formate as isotope source (Scheme [2]). Again, this method tolerates a high variety of functional groups like amines, amides, esters, carboxylic acids, alcohols, etc. and can thus be widely applied. It is therefore extending the toolbox of deuteration by another valuable addition.

Zoom Image
Scheme 2 Reductive deuteration of aromatic or unsaturated amines with deuterated ammonium formate. a Isolated overall yield of deuterated products [Dx ], chemical purity >95% (determined by NMR, HPLC or LC–MS). b Determined by MS analysis of a single experiment after filtration of the reaction mixture; replications of the experiments yielded slightly different mass distributions; [%D] value shows the relative distribution of mass units in the product.

The group of Sajiki published an impressive paper on direct multi-deuteration of aliphatic carboxylic acids via platinium catalysis.[7] As deuterium source D2O/i-PrOD-d 8 are used with a Pt/C catalyst and thermal activation at 120 °C. To obtain the desired deuterated product, a lot harsher conditions were used prior to this great new tool. Harsh conditions with >220 °C and 2.3 MPa D2 gas were necessary to reach the goal. Thus, this method quickly prooved superior and opened up new options for mild and cheap deuteration of aliphatic carboxylic acids, used for analytical and metabolism studies or the developemnt of deuterated drugs.

2.1.1

Labeled Nitrosamines – The Hunt to Quantify Hazardous Impurities

A great example to understand the impact of this application is the recently discovered nitrosamine case. Nitrosamines have proven cancerogenic properties and were found in several sartan drugs.[8] As response, both the EMA’s human medicines committee (CHMP) and the FDA requested a review process of nitrosamine quantification in chemically synthesized human medicines. Starting in March 2020, the health authorities ordered a risk assessment for all active pharmaceutical ingredients (APIs), based on the manufacturing process and identification of potential risks of nitrosamine impurities. Since then, highly accurate LC–MS or GC–MS methods are used by GMP-qualified analytical laboratories worldwide to quantify nitrosamine impurities in APIs. Here is also where deuterated compounds come into play: stable isotopically labeled analytical internal mass spectrometry (MS) standards of nitrosamines, as can be seen in Figure [2], are used to support and facilitate the analytical process. The measurements for accurate detection of potential nitrosamine impurities are essential and no API using secondary amines in the production process is allowed to be sold without it.

Zoom Image
Figure 2 Examples of deuterated nitrosamine MS standards for LC–MS impurity quantification

In a publication from 2022 we describe the synthesis and analysis of this compound class.[8a] The deuterated nitrosamines are obtained by alkylation of Boc-protected amines with isotopically labeled alkyliodides. After Boc deprotection the amine was transformed to the nitrosamine by treatment with sodium nitrite and sodium hydrogen sulfate. Analytic analysis will then show a shift in retention time of the deuterated compound and the expected mass pattern.


# 2.1.2

Deuterated Drugs, an Approach To Improve Existing Drugs or To Find Opportunities in Drug Discovery

Deuteration of drugs is arguably the most conservative form of bioisosterism, a process where a substructure is replaced to improve one or more properties of the original compound, while retaining its biological activity. In case of deuteration the change is minimal, as the two isotopes are very similar in terms of physicochemical properties. The one substantial difference is the twofold larger mass of D compared to H. The mass difference results in a reduced vibrational stretching frequency compared to C–H bond, which in turn leads to a lower ground-state energy and thus a greater activation energy for cleavage. Thus, the C–D bond is more stable than the C–H bond and its cleavage occurs in a slower fashion. In contrast to other metabolic blockers, such as halides, deuteration increases the resistance of a molecule to bond cleavage, without significantly altering its steric hindrance or electronic properties. Despite the small difference, deuteration can have a tremendous effect on a drugs property. The interest of deuteration within the drug hunter community increased significantly when the effect of this modification improved the API well beyond lipophilicity and pharmacokinetic (PK) properties, but also impacted the efficacy and safety profile. One example for this is deutetrabenazine, the first deuterated drug approved by the FDA in 2017. This analogue possesses a superior PK profile compared to the original drug, tetrabenazine, a substance that was approved in 2008 for the treatment of chorea associated with Huntington disease. Deutetrabenazine allowed a significant reduction in both dose and dose frequency, thus changed the patient’s life for the better. Since then, several deuterated drugs have made it to the market and a recent review covers this topic thoroughly.[9]

Zoom Image
Scheme 3 HIE reactions of Kerr catalyst with tetrazoles.

#
# 2.2

Tritium-Labeling Methods – The Fast Approach to Radioactively Labeled Compounds

In 2010, a shortage in 14C availability initiated a change in our strategy of radiolabeled compounds supply. We became interested in working with tritium gas with the focus of fast delivery procedures. In addition to state-of-the-art methods, such as tritiation of double bonds or halogen–tritium exchange, we also evaluated recently reported advances in hydrogen isotope exchange (HIE) reactions. Based on the earlier work from Heys, Hesk, Kerr,[54] and many others,[10] we were able to test several active catalysts to solve our synthetic challenges. Nevertheless, we observed all known methods did not help us to be reactive and fast enough to deliver all tritiated compounds in time to be useful for the drug research projects. Therefore, we became highly interested to develop new methods to overcome the challenges in tritiated compound supply, focusing on late-stage functionalization. While we have profited from the motivation and spirit of several master’s and bachelor’s students, as well as very talented PhD students with funding from the European Union in International Training Networks (ITN) programs Isotopics and Isobiotics, we have done most of our chemistry research projects in collaboration with academic partners.

2.2.1

Hydrogen Isotope Exchange by Iridium catalysis

In our first tritium chemistry research project we teamed up with the group of William J. Kerr to explore the effect of heterocycles such as oxazoles, imidazoles, or thiazoles as directing groups in HIE reactions. Later on, we expanded the utilization of Kerr’s catalyst to tetrazoles,[11] finalizing the project by tritiation of valsartan in a single reaction step (Scheme [3]).

Since iridium catalysts, such as those explored by Crabtree,[12] Kerr,[13] Pfaltz,[14] Burgess,[15] and Ding,[16] are commercially available, they can be easily tested in HIE reactions by an easy reaction setup and analytical control (Figure [3]). HIE reactions can be easily followed by combination of mass spectrometry (LC–MS or GC–MS) and 1H or 2H NMR spectroscopy. Unselective HIE methods generally result in a broader mass distribution due to a high number of isotopologues. 1H NMR spectroscopy can be used for quantifying and determining which positions underwent HIE because the parent and deuterated compounds NMR spectra are compared. The missing signals and the altered ratio of integrations from the parent compound indicate the deuterium incorporation. After a parallel screening setup using D2, the best reaction conditions are transferred into a deuterium manifold, where low deuterium pressures (25–200 mbar) are evaluated to optimize the conditions for the final tritium reaction. This is necessary as the kinetic isotope effect for hydrogen isotopes is the highest in the periodic table. The mass difference from hydrogen to deuterium or hydrogen to tritium is either 200% or 300%, respectively. Another critical parameter is the scale (1–5 mg), as only low amounts of radioactive gas are preferably used.

Zoom Image
Figure 3 Reaction set-up and analytics of deuterium and tritium reactions
Zoom Image
Figure 4 Graphical overview in which temperature window the iridium catalysts are efficiently working (threshold 25%D) for HIE reactions with acetamidoacetophenone; the yellow star is indicating the highest deuterium incorporation

In HIE reactions of 4-acetamidoacetophenone with several iridium catalysts in dichloromethane at low to higher temperatures (–80 °C to 130 °C) we have studied the differences of several iridium catalysts[17] (Figure [4]). Interestingly, depending on the catalyst we observed strong differences in reactivities and selectivity of the deuterated product. Generally, catalyst screenings are routinely done in industry to optimize challenging reactions, however, only sparsely reported from academic groups. These studies can be very valuable as they give an overview about reactivity, selectivity, and stability of the studied catalyst or reagent.

Zoom Image
Scheme 4 HIE reaction of amino acids and peptides with Kerr catalyst; 31 Ci/mmol = 1.15 TBq/mmol.
2.2.1.1

HIE with Iridium(I) Kerr’s Catalysts

Consequently, Kerr’s catalyst became one of our standard catalysts in early HIE method screening. Originating from such a screen we discovered the unusual HIE reaction of DM4 (Scheme [4]). In contrast to our prior predicted aromatic deuteration position we obtained the hydrogen exchange in the aliphatic position of the DM4 side chain. After further elaboration in the critical parameters of the aliphatic HIE (namely these are solvents and reaction temperature as most important), we were able to extend the methodology to protected amino acids, predominantly glycine and alanine. In the case of alanine, the HIE reaction took place with retention of the stereoinformation, selectively at the α proton of the amino acid. Finally, we were able to transfer the chemistry to reactions with tritium. Notably, reactions with tritium are performed with 3–10 equiv of tritium at low pressure (25–300 mbar). Even though the reaction conditions are optimized with deuterium gas, the exact kinetic of these reactions are still difficult to predict and sometimes higher catalyst loadings are utilized to circumvent overnight reactions. However, the use of 50 mol% catalyst like in this case is an exception, normally the reactions are optimized to apply 2–10 mol% catalyst.[18]

In the same year Kerr in collaboration with us reported the directed and selective HIE at sp3 C–H centers, resulting in high levels of D incorporation with low catalyst loadings and under mild reaction conditions applying Kerr’s catalyst (Scheme [5]).[19] A broad variety of substrates such as heterocycle-substituted morpholines, piperidines, piperazines, azepine, pyrrolidine, or azetidines were explored. With the heteroatom in the aromatic substituents acting as directing groups for the catalyst, the protocols proved to be efficient and reliable across a wide range of saturated heterocycles and aliphatic units. The antidepressant mirtazapine does not contain any sp2 centers which could be labeled via directed HIE; however, its pyridine nitrogen could direct sp3 labeling. We observed a high D incorporation of 94% on the piperazine ring. Similarly, a high incorporation was obtained on the piperazine ring of the tranquilizer azaperone and notably with excellent selectivity versus the ortho sp2 exchange directed by the ketone. Finally, the stimulant caffeine reacted in a highly selective deuteration at the 7-methyl position, directed by the imidazole nitrogen.

Zoom Image
Scheme 5 Deuteration of a variety of heterocycle-substituted tertiary amines with the Kerr catalyst
Zoom Image
Figure 5 Improved radiochemical yield for tritium reactions. Literature analysis over the last 20 years; 20 Ci/mmol = 0.74 TBq/mmol.

Both examples showed that methodology development with an already established catalyst can lead to fascinating new HIE applications and unexpected reactivity.

Another topic highly important in the industrial context and for chemists in general is to increase the sustainability and environmental footprint. Therefore, we became interested to study possibilities to decrease the radioactive waste. Typically, tritiation levels delivered in standard reactions far exceed the levels of radioactivity required for subsequent studies and thus even low-yielding transformations may provide enough quantities of the desired tracers. If reported, radiochemical yields for tritiation reactions are typically below 10% with only very limited examples exceeding this threshold (Figure [5]). With this backdrop, we felt that enhanced focus on the radiochemical yield (RCY) relating to reactions involving tritium gas is merited. In our proof-of-concept reactions with our optimized HIE reactions with complex drug molecules, we were able to proof the potential to significantly increase the RCY and by this also to reduce radioactive waste. In the reaction with fenofibrate we obtained the isotopic-enriched products with 1.0 tritium atom corresponding to 47% RCY. We reacted dolasetron (serotonin 5-HT3 receptor antagonist used to treat nausea and vomiting following chemotherapy), celecoxib (COX-2 inhibitor and nonsteroidal anti-inflammatory drug (NSAID) to treat rheumatoid arthritis) and paracetamol (medication used to treat pain and fever) by the same procedure to demonstrate the broader scope on real drug applications. To our delight we isolated all tritiated compounds with RCY in a range of 13–47%, which is a significant improvement (3–10 times higher) to the RCY values reported so far. We also compared our RCY-optimized conditions with the state-of-the-art HIE protocol applying 10 equiv of tritium gas. As could be expected the specific activity increased from 20–30 to 46–95 Ci/mmol (1.7–3.5 TBq/mmol) if more excess of tritium was used, however, the RCY dropped significantly from 32–47% to 6–11%. Besides, the specific molar activity, mostly the parameter to justify high excess of tritium, ranged from 15–30 Ci/mmol (0.6–1.1 TBq/mmol) which is suitable for most in vitro application experiments.[20]

In contrast to carbon-14, it is widely accepted in tritium chemistry simply to use a large access of tritium gas without optimizing conditions for tritium consumption at all. We have introduced the concept of a new optimization parameter with the DCY (deuteration chemical yield) values to optimize tritium gas reactions in a nonradioactive reaction setup. By optimizing the DCY it enables academic or industry scientists without radioactive working space to participate in the challenge to further improve the RCY in tritium reactions. We envision to change the evaluation level of successful tritium reactions and to care more about radioactive resources to decrease the burden of radioactive emissions in general. Nevertheless, we are convinced that our results are only the first step, and further improvements of catalysts and understanding of HIE mechanism are necessary in the future.

The direct labeling of large biomolecules, such as antibodies, can be developed as alternative approach. However, the challenge of performing the labeling procedures in water or buffers needs to be overcome. A first step in this direction was recently described by us (Scheme [6]). We developed a HIE reaction with water-soluble Kerr-type iridium catalysts and demonstrated successful conditions with several buffers and in a pH range from 5–12. The reaction conditions could be applied to tritium chemistry. DFT calculations provided a rationale which directing groups could be applied with the prepared catalyst under aqueous conditions.[21]

Zoom Image
Scheme 6 Hydrogen isotope exchange by homogeneous iridium(I) catalysis in aqueous buffers with deuterium or tritium gas

Finally, beside the search to expand the Kerr-catalyzed HIE reaction to new substrates, new reactivities, challenging solvents, or optimized RCYs, we also tried to increase our understanding by comparison of experimental and computational data. By this endeavor we have found a clear order of the influence of the directing group in HIE reactions with Kerr’s catalyst. These results allow predictions in regioselective HIE reactions of complex molecules and give a perspective to the concept of directing-group-induced late-stage functionalization in general. It is reported that the rate-determining step and the free activation energy of Ir–CH insertion explain the formation of the deuterated products. However, the order of directing-group significance in the studied compounds differed compared with the order observed experimentally in HIE competition experiments. It was not the rate-limiting transition state that was responsible for the HIE reaction outcome in competition cases, but the lower free energy of the initially formed coordination complex of substrate and HIE catalyst. With this concept we aimed to improve the outcome of computer-aided predictions of HIE reactions and potentially increase the effectiveness of late-stage functionalization reactions in research. We have highly profited from the work reported by Kerr and Tuttle to generate a better understanding of the HIE reaction mechanism and the energetic profiles in the transition states. They have further developed the field to develop computer-aided catalyst design to predict HIE reaction outcome.[22]

Zoom Image
Scheme 7 Deuteration with Tamm’s catalyst on anilines

# 2.2.1.2

HIE Reactions with Other Iridium(I) Catalysts

In our efforts to increase our armory of HIE catalysts, we initiated a collaboration with Prof. Matthias Tamm. He had published several novel iridium catalysts and we agreed that it would be highly interesting to study them in HIE reactions. Therefore, a new class of bidentate RP,NR′ ligands, bearing the electron-rich imidazolin-2-imine group as N donor, were used for the synthesis of iridium(I) complexes.[23] The complex [(t-BuP,NMe) Ir(COD)]BArF24 (COD = 1,5 cyclooctadiene) was found to be particularly active in the initial deuteration study of a total of 23 substrates. The substrate scope included examples with common directing groups such as acetyl, heterocycles, sulfones, nitro groups, and benzylamines, in addition to unique directing groups such as the Boc protecting group in anilines and other aromatic amines or the methoxy moiety. Later, the catalyst became known as Tamm’s catalyst (Scheme [7]).

Together with Tamm, we investigated a catalytic protocol for a highly selective HIE of phenylacetic acid esters and amides under very mild reaction conditions.[24] Using Tamm’s catalyst, the HIE reaction on a series of phenylacetic acid derivatives proceeded with high yields, high selectivity, and with deuterium incorporation up to 99%. The method was fully adaptable to the specific requirements of tritium chemistry, and its effectiveness was demonstrated by direct tritium labeling of the fungicide benalaxyl and the drug camylofine (Scheme [8]). Further insights into the mechanism of the HIE reaction with Tamm’s catalyst were provided utilizing DFT calculations, NMR studies, and X-ray diffraction analysis. The results were remarkable as iridium-catalyzed HIE processes had only been successful so far for 5-mmi transition states or in rare cases via a conjugated 6-mmi transition state such in acetanilides (mmi = mass/moment of inertia). Aromatic compounds with a CH2-benzylic bridge remained a challenge and were either deuterated via palladium catalysis in deuterated acetic acid by Yu or in a two-step ligated-palladium-substrate complex.[25] The later one was successfully transferred to conditions for tritium chemistry by Hoover.[26] The new application of Tamm’s catalyst closed this gap conveniently to perform tritiations in a single reaction step at room temperature via a 6-mmi transition state.

Another internal drug research project triggered the methodology development of HIE of sulfonylureas. As all standard methods in our laboratory failed and, in the literature, only one example of a secondary sulfonamide was reported, we expanded our method screening to nonexplored catalytic, commercially available systems. By this approach we discovered the Burgess catalyst, mostly known from asymmetric hydrogenations, is active in a HIE setting with chlorobenzene solvent at 120 °C.[27] We expanded the scope and finally we reported the development of the first practical HIE protocol for the deuteriation of various secondary and tertiary sulfonamides as well as sulfonylureas (Figure [6]). The method had a broad substrate scope and was applied to sulfa drugs, such as glibenclamide. Another example are the different labelling opportunities shown for apixaban (Figure [6], bottom). While applying the Spinphox iridium catalyst deuteration was only found in the ortho position (position A) of the cyclic lactam moiety, with Burgess catalyst unselectively three benzene positions (A, B, C) were exchanged without high discrimination. Depending on the chosen catalyst and conditions significant differences in deuteration outcomes were observed.

Zoom Image
Scheme 8 Calculation of transition state and further understanding of Tamm’s catalyst behavior (14 Ci/mmol = 518 GBq/mmol)
Zoom Image
Figure 6 Deuteration with Burgess catalyst (2.3 Ci = 85 GBq; 9 Ci/mmol = 333 GBq/mmol).

These screening findings often led to further small research programs, exploring for example the applicability of Burgess or Ding’s Spinphox catalyst in HIE reactions.[28]


#
#

2.2.2 Ruthenium-Catalyzed HIE

Together with Lutz Ackermann, we were interested to elaborate a method for tritium labeling of carboxylic acids. We developed well-defined ruthenium(II)biscarboxylate complexes and performed selective ortho deuterations with weakly coordinating carboxylic acid with outstanding levels of isotopic labeling. The catalytic system was highly stable with a broad functional group tolerance in an operationally simple manner, allowing the isotope labeling of challenging pharmaceuticals and bioactive heterocyclic motifs. The synthetic power of our method was highlighted by the selective tritium labeling of repaglinide, an antidiabetic drug, providing access to defined tritium-labeled therapeutics (Scheme [9]). The respective tritium labeling was performed in our labs using freshly prepared T2O from PtO2 and tritium gas.[29]

Zoom Image
Scheme 9 Ruthenium-catalyzed HIE reaction of benzoic acid derivatives (5.5 Ci/mmol = 0.2 TBq/mmol)

# 2.2.3

Nanoparticles as Catalysts in HIE

In the EU funded ITN Isotopics we developed together with Bruno Chaudret and Gregory Pieters the HIE protocol of aniline deuterium and tritium incorporation with ligated iridium nanoparticles. The reaction of the precursor [(COD)Ir(MeO)]2 with 5 bar of H2 over in the presence of N-heterocyclic carbene (NHC) ligand ICy (1,3-dicyclohexylimidazole) produced very small iridium nanoparticles (IrNPs) of 1.1–1.3 nm mean size as observed by TEM analysis (Scheme [10]). While the scope of the novel catalytic HIE system was elaborated with deuterium, the target was the final application in tritium labeling with T2 gas. The use of tritium gas as isotope source is a clear advantage because it is the cheapest and easiest raw material to handle for tritium labeling nowadays. Typically, reactions involving T2 gas are conducted using a subatmospheric pressure of T2 to minimize the risk of leakage and radioactivity release. Therefore, we optimized our reaction conditions using volixibat aniline pharmacophore as substrate. Using a catalytic loading of 80 mol% of IrNPs, the tritiated compound was obtained utilizing subatmospheric pressure of tritium gas (p = 0.8 bar, 10 Ci = 370 GBq) within 3 h. In terms of regioselectivity, the labeling of the less sterically encumbered ortho position of the NH2 group was mainly observed (in good congruence of the deuterium incorporation observed for D2 experiments) and analyzed by 3H NMR spectroscopy. The high specific activity of 25 Ci/mmol (0.9 TBq/mmol) obtained clearly demonstrated the high potential of this method (Scheme [10]).[30]

Zoom Image
Scheme 10 Nanoparticle formation and heterogeneous HIE catalysis of anilines; IrNP = iridium nanoparticles

Lead by Gregory Pieters, we have helped to demonstrate that [Ir(COD)(OMe)]2, a commercially available and air-stable Ir dimer, can be used as precatalyst to generate in situ Ir complexes and nanoclusters under a D2 or T2 atmosphere (Scheme [11]).[31] Compared to state-of-the-art methods for deuterium or tritium labeling, this approach possesses several key advantages: a) the use of an easy-to-handle, air-stable, and commercially available catalyst; b) a very broad substrate scope allowing the labeling of recurrent substructures in pharmaceuticals such as azines, indoles, carbazoles, oxa- and thiazoles, thiophenes, and nucleobases; c) a high functional group tolerance with a compatibility of halogens such as Cl, Br, F, and nitrile-containing compounds; d) hydrogen isotope incorporation of multiple carbon positions allowing for one-step synthesis of complex pharmaceuticals with high deuterium contents or very high molar activities (up to 122 Ci/mmol = 4.5 TBq/mmol). All these features, as well as the easy implementation of this method in normal equipped laboratories, can facilitate the synthetic access to deuterated/tritiated analogues of complex molecules, which are essential diagnostic tools in drug discovery and development.

Zoom Image
Scheme 11 Ir nanoparticle assembly using D2 or T2

# 2.2.4

Photoredox-Catalyzed HIE

A breakthrough towards a simplified and easy-to-use HIE protocol, using photoredox-mediated activation,[32] was reported by MacMillan in 2017 exploring the deuteration and tritiation of complex tertiary amines at the α-position to the nitrogen with good to high selectivity.[33] We became highly interested to apply this new methodology to our daily challenges. Inspired by this method, we developed a photoredox-catalyzed HIE protocol to label lysine or ornithine with deuterium or tritium.[34] We demonstrated that by selecting the right combination of high pK b organic base, thiol, and photocatalyst, HIE reactions with amino acids and peptides, which use D2O as the deuterium source, can be achieved under simple blue-light irradiation (λ = 450 nm; Scheme [12]). For various protected small peptides, containing between 3–8 amino acids, the reaction was found to be highly selective for the α-position of the primary amine located on the lysine side chain, and tolerant to many functional groups. The reaction was successfully applied to the deuteration of many peptide-derived drugs, achieving similar results with both simple and more complex structures such as icatibant. Finally, we tried to transfer the method to the tritium chemistry. Unfortunately, we were only able to prove the applicability with our model compound lysine, more complex compounds failed due to challenges in the downscaling process.

Zoom Image
Scheme 12 Photochemical deuteration and tritiation of peptides; photocatalyst 2 = Ir[dF(Me)], BTMG = Barton’s base

Burkhard König developed a practically simple and sustainable process for the site-selective carbanion generation from C(sp3)–H bonds in the α-amine position (Scheme [13]). We helped applying this methodology to numerous, druglike amines which were selectively deuterated/tritiated with a high degree of regioselectivity. In addition, the methodology was used for the incorporation of the 1,2-amino alcohol motif in various molecules, including the drug nabumetone. The practical utility of this method was demonstrated by conducting synthesis on gram scale followed by amine deprotection. Based on mechanistic experiments, including spectroscopic analysis, a radical-polar crossover mechanism was proposed. The amine substrate was oxidized by the excited photocatalyst, leading to an α-amino radical after deprotonation. This intermediate was then reduced by the radical anion species of the photocatalyst to form the desired carbanion, facilitated by the electron-withdrawing ester group. Interestingly, an aromatic protecting group (e.g., para-methoxyphenol (PMP)) on the amine was required, whereas analogous amines with purely aliphatic substitutes do not undergo this reaction. The selectivity is hence driven by both, the PMP and the glycyl moiety, which was exploited for the challenging regioselective labeling of a molecule containing two distinct amino ester groups. It is worth mentioning that untargeted chiral centers stay untouched, while a classical approach of employing a strong base is expected to lead to racemization. Finally, we showcased the applicability of the protocol for the site-selective tritiation of a linezolid derivative. The tritiated product was obtained in high purity after HPLC purification with a good tritium incorporation 1.2 T/molecule (36 Ci/mmol = 1.3 TBq/mmol).[35]

Zoom Image
Scheme 13 Photochemical deuteration and tritiation using D2O and T2O; 3DPA2FBN = 2,4,6-tris(diphenylamino)-3,5-difluorobenzonitrile

Even though these methods have proven their applicability, the production and handling of T2O remains a challenge. The first ones overcoming this problem were Yang and Lehnherr from Merck (USA). They reported the application of homogeneous rhodium catalysts (Wilkinson’s catalyst (RhCl(PPh3)3) as a precatalyst for activation of hydrogen gas in a photoredox-catalyzed HIE reaction to switch from T2O to T2 as an isotope source.[36]

We became interested in exploring the photoredox-catalyzed HIE reaction with tritium gas utilizing heterogenous, in situ prepared metal nanoparticles as hydrogen atom transfer (HAT) precatalysts. We initiated our studies by screening several heterogenous catalysts in the photoredox-catalyzed HIE reaction of clomipramine as a model compound (Scheme [14]). We tested several precatalysts with 4CzIPN as the photocatalyst for a possible starting point. To our delight, we found several conditions promoting a photocatalyzed HIE with in situ generated metal nanoparticles from homogeneous iridium or rhodium precursors.[37]

We observed synergies of two possible HIE pathways via photoredox catalysis or CH functionalization processes. In total, we have expanded the scope of photocatalyzed HIE reaction by applying in situ generated iridium nanoparticles as HAT precatalysts with deuterium or tritium gas. By using the combination of the photoredox-catalyzed radical pathway as the CH activation and functionalization at the activated nanoparticle surface, high specific activities for complex molecules can be easily achieved under mild conditions. This was the first example of HIE reactions catalyzed by a combination of heterogeneous catalysis and photochemistry. The method adds to the armory of simply handled tritiation methods applying tritium gas as isotope source.


# 2.2.5

HIE via Classical Radical Mechanism

Studer developed a robust metal-free remote HIE reaction at various C(sp3)–H bonds for δ-deuteration of primary amines, γ-deuteration of aliphatic acids, and α-deuteration of secondary amines. For the α-deuteration of secondary amines, an easily removable β-alanine-based auxiliary was used.[38]

Zoom Image
Scheme 14 Combined photo- and metal-catalyzed deuteration of complex molecules, NHP1 = Ir2Cl2(COD)2, PC1= 4CzIPN (1,2,3,5-Tetrakis(carbazol-9-yl)-4,6-dicyanobenzene, 2,4,5,6-tetrakis(9H-carbazol-9-yl) isophthalonitrile)

The operationally simple and mild method allowed for highly regioselective C–H deuteration mediated by intramolecular 1,5- or 1,6-HAT. Various nonactivated and activated C(sp3)–H bonds as well as primary C–H bonds next to heteroatoms were monodeuterated site-selectively in good yields and with high deuterium incorporation. The scope was demonstrated by the successful monodeuteration of more complex compounds, including natural product or pharmaceutically relevant compounds (Scheme [15]). Even though the method was highly convenient for deuterium incorporation, the transfer to tritium yielded only medium successful results. In a single case the radioactive compound was isolated with 0.3 %T and a specific activity of 1 Ci/mmol (= 37 GBq/mmol).

Zoom Image
Scheme 15 Radical-mediated deuteration and tritiation of complex molecules.
Zoom Image
Scheme 16 Synthesis of tritiated taxane azide

# 2.2.6

Beyond HIE – Halogen–Tritium Exchange

Even though the examples for successful HIE are impressive to outsiders it generates the wrong impression that already >50% of all tritium projects are solved in this way. The truth is, a significant part (+/– 30%) is still applying classical approaches such as palladium-catalyzed halogen–tritium exchange. As example we want to discuss the preparation of tritium-labeled photoaffinity probes of two cancer-treatment drugs (Scheme [16]). Both syntheses started from their diiodo precursors and were supposed to be performed via iodine–tritium exchange with tritium gas. However, this standard approach turned out to be a unique challenge as either the reaction was not complete and only the monotritiated compound was detected or by prolongation of the reaction time overreduction of the taxanes to the corresponding 3H-anilines were observed. The successful tritium labeling of both compounds was finally achieved by addition of a ‘sacrificial’ azide to prevent overreduction of the taxane azidophenyl moiety.[39] Excess of tritium reacted with the victim azide leaving the taxane azide untouched. By this protocol we obtained the desired material with specific activities of 51 and 55 Ci mmol–1( 1.9 and 2.0 TBq mmol–1), respectively, which was sufficient for their use in Pgp transporter study experiments.[40]


#
#
# 3

Challenges in 14C-Synthesis Projects

Carbon-14 is the naturally occurring, radioactive isotope of carbon-12. It has a half-life of 5730 years and emits low-energy β-particle radiation, which allows tracking of this label by β-counting or β-imaging technologies. As carbon labels are a lot more stable than hydrogen labels, 14C has a significant advantage compared to tritium.[41] It is due to its chemical stability that 14C is commonly used for in vivo ADME and PD studies. Also, for first in human experiments 14C is the label of choice. Its superior stability comes with a price though, as the installation of the label mostly includes a multistep synthesis and high costs of starting material. Commonly C1 building blocks like 14CO2, 14CO, or 14CN, are used to label a drug, making it necessary to set up a whole new synthetic route for the labeled drug candidate. The challenge of working with stochiometric amounts of radioactive gases has opened the need for a completely new methodology to effectively use the expensive and rare material. While the multistep approach and late-stage carbon-isotope labeling is still dominating the labeling field, the new development of carbon–isotope exchange reactions promises a fundamental paradigm change.[42]

Effective, optimized, and high-yielding reactions are key to a good labeling approach. One successful example for a multistep approach is the synthesis of SGLT inhibitors AVE2268 and AVE8887 (Figure [7]).[43]

Zoom Image
Figure 7 SGLT inhibitors AVE2268 and AVE8887

Despite their similarity AVE2268 and AVE8887 had to be labeled in two different routes. For AVE2268, a five-step synthesis was used. The synthetic route started off with a Friedel–Crafts acylation of 3-methoxythiophen, followed by deprotection of the methoxy group with boron tribromide. The hydroxy group was alkylated with acetobrom-α glucose. Subsequently the ketone was reduced using sodium cyanoborohydride and TMS-Cl, and in a last step the protection groups were removed under basic conditions. The desired product was obtained in 50% overall yield (Scheme [17]).[43]

Zoom Image
Scheme 17 14C-labeling route to SGLT inhibitor AVE2268
Zoom Image
Scheme 18 14C-labeling route of AVE0991

Unfortunately, this route was not applicable to AVE8887 and a new route was developed. The synthesis starts with 14C trifluoromethoxybenzoic acid and is converted into a Weinreb amide under Appel conditions. Addition of the lithiated thiophene gave the corresponding ketone, which was deprotected and alkylated according to the previous synthetic route. This six-step procedure gave the desired product AVE8887 in 45% overall yield.[43]

Both synthetic routes are linear, high yielding, and with several radioactive steps. A more parallel approach is shown in our next example, the radiosynthesis of 14C-AVE0991 (Scheme [18]).[44] Within this route two building blocks were synthesized and then coupled, which reduced the radioactive steps for this complex molecule to 4 and gave an overall radiochemical yield of 21%. The synthesis starts by alkylation of an imidazole derivative with 4-bromo-14C-benzyl bromide. This building block is coupled with the second nonradioactive building block in a Stille reaction. After isocyanate coupling and methoxylation the final product is obtained.

Zoom Image
Scheme 19 Catalytic cycle of the carbon isotope exchange (CIE) of amino acids; RCY: radiochemical yield, red/blue: isotope incorporation in %, yield in brackets. a TE: trapping efficiency.

In a recent collaboration with Prof. Lundgrens group a new procedure of 14C labeling was explored via carbon isotope exchange (CIE).[45] The new method to label amino acids in a one-step procedure with CO2 was published in Nature Chemistry in 2022.[46]

The reaction starts with condensation of an aldehyde with the amine of the amino acid, this leads to a more dynamic keto–enol tautomerism of the acid, which can be attacked by CO2 when in its enol form. The obtained malonic acid derivative reacts by decarboxylation of one carboxylic group giving a 1:1 chance of eliminating the unlabeled group. As the reaction can undergo several carboxylation/decarboxylation cycles, the ratio of labeled to unlabeled CO2 is critical for specific activity of the final product. Despite other parameters, for radiochemists the radiochemical yield is the most important result, which means stochiometric amounts of CO2 give better results than excess use of radioactive gases. Optimized reactions gave the best results with 2.6 equiv of 14CO2, resulting in a radiochemical yield of 11% and 53% 14C incorporated into the amino acid (Scheme [19]).

With more and more CIE reactions being published, new methods conquer the synthesis field and making easier and faster production of labeled material accessible. This new field is highly interesting and should be followed closely.[47]


# 4

Short-Lived Isotopes – The Need for Speed

Next to rather long-living radioisotopes like 14C and 3H, there are plenty of short-lived isotopes, which deserve our attention as well. Due to their short half-life, those radioisotopes do not generate the problem of radioactive waste and do not cause long-term radiation for in vivo experiments. Thus, they are preferably used in diagnostic. One tool is positron emission tomography (PET) imaging. For this analytic tool, radioisotopes that undergo β+ decay are required. The emitted positron travels a specific distance (depending on the average β+ emission energy of the isotope) in the tissue until it lost enough kinetic energy and reacts with an electron. Thus, photons are released that can be detected by a scintillator in a scanning device and transferred into a measurable signal. Commonly used is 18F, as it has a comparably long half-life with 110 min, compared to other PET-compatible isotopes like 11C (t1/2 = 20 min), 13N (t1/2 = 10 min), or 68Ga (t1/2 = 67 min). This gives 18F-labeled compounds enough time for their synthesis and purification. In addition, its short positron linear range in tissues (2.3 mm) allows the highest resolution in PET images of all available positron emitters. One example of a 18F-labeled drug for PET imaging is our work on 18F-TAK875 (Figure [8]), which was used as β-cell imaging probe, addressing the free fatty acid receptor (FFAR1/GPR40).[48] In a total synthesis time of 60–70 min of the labeling step, the tracer was made in a 17% radiochemical yield. For the installation of the 18F label a TAK875 derivative was synthesized in which the methyl on the sulfonyl group was replaced by tosyl-propanol. 18F-labeling of the tosylate was then performed using azeotropically dried [18F]-fluoride, which was activated by a potassium carbonate/K2.2.2 system.[42] The analytics showed successful results for the labeled compound, and it was further elaborated to an automated radiosynthesis for its use in PET imaging.

Zoom Image
Figure 8 Structure of 19F/18F-TAK875

Another interesting short-living isotope is 68Ga. It has a half-life of only 68 min, but is commonly used in PET studies, helping as diagnostic tool to bring the right medicine to the right patient. Isotopes with a short half-life have a higher specific activity and are easily traced even in whole-body analyses. Our team worked out a method to use these properties for a new diagnostic tool for nonalcoholic hepatosteatosis (NASH) and fibrosis.[49] A specific bifunctional binder for the hepatocyte specific ASGPR receptor was prepared in a synthetic route with more than 20 steps. It was designed to study the receptor decrease during development of liver fibrosis in a rat model of NASH by fluorescence tomography and PET imaging. In more detail, a bifunctional triantennary GalNAc warhead was combined with 68Ga-labeled dodecane tetraacetic acid (DOTA) for PE imaging and a heptamethine cyanine dye to allow near-infrared optical imaging of the construct. The approach proved successful in liver-selective determination and quantification of gradual changes of ASGPR receptor during NASH progression in our ZSF1 rat model, a model that mirrors the obesity and insulin-resistant characteristics of the disease in humans. This new method takes us another step closer to understanding and exploring a disease that impacts the lives of millions of patients by helping to understand the impact of treatment intervention in a more relevant in vivo model, and thus providing a better translation of preclinical research into the clinical setting.


# 5

Beyond Isotope Science – Late-Stage Functionalization

The introduction of a radioactive isotopic label into a molecule generally follows the idea to depend on as little as possible of radioactive synthetic steps. This approach decreases costs, time, but most of all radioactive waste. Therefore, the concept of late-stage functionalization is a principal methodology applied in isotope sciences. As was already described in the paragraphs before the methodology development, especially in the areas of isotope exchange (hydrogen and carbon), has progressed significantly. The learnings of these concepts can be utilized in other areas of applied chemistry such as lead optimization of compounds.

5.1

Examples of Late-Stage Functionalization for Peptides

Chemoselective functionalization of peptides and proteins to selectively introduce residues for detection, capture, or specific derivatization is of high interest to the synthetic community. We have therefore developed a method for the mild and effective monoiodination of tyrosine residues in fully unprotected peptides. This method is highly chemoselective and compatible with a wide variety of functional groups. The method tolerates other aromatic amino acids such as tryptophane and histidine, but also sulfur-containing amino acids like methionine or cysteine. By exploiting the reactivity of the monoiodo peptides we performed conjugation with fluorescent building blocks via Suzuki–Miyaura cross-coupling, as an example of bioimaging probe synthesis. Furthermore, the iodinated peptides can be used for chemoselective labeling of peptides with tritium in an iodine–tritium exchange.[50] All these selective functionalizations of complex molecules are generally known as late-stage functionalizations.[51]

In another research project we helped to developed together with Olga Mancheno a visible-light-mediated C–C bond functionalization of electron-rich benzylic C–H bonds in phenolic ethers or tyrosine derivatives (Scheme [20]). A broad range of valuable functional groups could be selectively introduced into the benzylic position of phenolic ether derivative backbones. The combination of gentle reaction conditions, chemoselectivity, and functional group tolerance was highlighted by the functionalization of complex natural products and drug building blocks. Hence, the demonstrated mild and selective generation of benzyl radicals in complex molecule scaffolds provides a powerful strategy with broad applicability in synthetic and medicinal chemistry.[52]

Zoom Image
Scheme 20 Late-stage functionalization of peptides: iodination of tyrosines (top); benzylic C–C coupling of tyrosines or phenolic ethers (bottom)

# 5.2

Examples of Catalyst-Controlled Late-Stage Functionalization

While control in late-stage functionalization can be obtained either by special reagents reacting with an activated position in the complex molecule such as in the examples mentioned above, we became also interested in studying catalyst-controlled reactions. In a collaboration with Huw Davies, we elaborated a method for site-selective C–H functionalization of N-(hetero)aryl piperidines, morpholines, and piperazines (Scheme [21]). The site-selective C–H functionalization of challenging substrates like N-aryl- and N-heteroaryl piperidines were achieved through chiral rhodium carbene intermediates, leading to the formation of highly stereoselective C-2 products. The transformation occurred regioselectivity at the α C–H bond next to the nitrogen atom, even though the β-heteroatom bond in morpholines and piperazines would deactivate this position. The reaction of piperidines delivered highly stereoselective outcomes. Extension of the method, however, to morpholines and piperazines furnished lower levels of stereoselectivities with the systems tested. It illustrated the potential of C–H functionalization by donor/acceptor carbenes and demonstrated their potential for rapid synthesis of pharmaceutically relevant chiral scaffolds.[53]

Zoom Image
Scheme 21 Site-selective C–H functionalization of N-(hetero)aryl piperidines, morpholines, and piperazines.

#
# 6

Conclusion

We have demonstrated how versatile and fascinating isotope science can be. The challenges are numerous, either in chemistry or isotope applications. For chemistry, even though some advances have been made, there are remaining many such as improvement of selectivity and reactivity in late-stage functionalization, low-pressure hydrogenation reactions, or carbon isotope exchange. These challenges can only be overcome in collaboration with extremely talented scientists in academia or industry in real teamwork.


#
#

Conflict of Interest

Anna Sib and Volker Derdau are employees of Sanofi and may hold shares and options of the company.

Acknowledgment

We would like to thank Anika Tarasewicz for proof reading of the manuscript and all current and former members of the Isotope Chemistry team Sanofi who contributed significantly over the years: Remo Weck, Mégane Valero, Jens Atzrodt, Martin Sandvoss, Claudia Loewe, Christian Klaus, Reiner Simonis, Anja Bartsch, Niels Griesang, Armin Bauer, Silvia Weber, Irmtraud Parker, Ingo Klein, Horst Medem, Raymond Oekonomopulos, Wolfgang Holla, Gerald Scholz, Andreas Weilbächer, Klaus Weihl, Klaus Döbrich, Monika Weber, Pia Ochsenschläger, Holger Thiel, Linda Raue, Dirk Gretzke, Stefan Raddatz, Jochen Zimmermann, Thorsten Fey, Gemma Solduga Ramirez, Sara Jobi, Volker Schaffnit, Lena Bendig, Kathleen Growe, Carsten Kroll, Johanna Zierow, Anja Dietrich, Carolin Schaab, Francois Brückner, Anne Christine Soldat, Juliane Sabine Wagner, Alexandra Paulick, Carina Bühler, Monja Gimber, Tanja Klingler, Julia Krebs, Kristina Thomas, Etienne Schmitt, Kevin Hörr, Mona Franz, Romain Bertrand, Patrick Rojahn, Marc Reid, Seth Jones, Mario Toller, Richard Mudd, Annina Burhop, Raphail-Robert Prohaska, Kristof Jess, Jennifer Blass, Anurag Mishra, Daniel Becker, Thomas Kruissink, Rebekka Weber, Christoph Hofmann, Valentin Müller, Fabien Legros, Viktor Pfeifer, Antonio Del Vecchio, Lin Wang, Tobias Brandhofer, Charlie Lim, Patrick Morawiecz, Lea-Marie Wagner, Ainara Bereziartua Unanue, Matea Srsen, Tobias Morgenstern, Karsten Donaubauer, Sara Kopf, Clara Marie Stork, Martin Stinglhamer, Michael Doyle, Daniela Schultheis, Korkit Korvorapun, Henrik Kramp, Marie Spiller, Elisa Martinelli, Marc San Jose Gracia. Furthermore, we would like to thank our collaboration partners, especially Prof. Matthias Beller, Prof. Matthias Tamm, Prof. William J. Kerr, Prof. Burkhard König, Prof. Armido Studer, Prof. Olga Garcia Mancheno, Prof. Huw Davies, Prof. Rylan Lundgren, Prof. Lutz Ackermann, Dr. Gregory Pieters, Prof. Bruno Chaudret, Prof. Daniele Leonori, and Prof. Siegfried Waldvogel.


Corresponding Author

Volker Derdau
Sanofi-Aventis Deutschland GmbH, Research & Development, Integrated Drug Discovery, Isotope Chemistry
Industriepark Höchst, G876, 65926 Frankfurt
Germany   

Publication History

Received: 31 October 2023

Accepted after revision: 04 December 2023

Accepted Manuscript online:
04 December 2023

Article published online:
17 January 2024

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


   Permissions and Reprints All articles of this category
Zoom Image
Zoom Image
Zoom Image
Figure 1 Areas of interest in an isotope chemistry pharma industry group by example of Sanofi
Zoom Image
Scheme 1 Mechanism of α,β-H–D exchange in alkylamines
Zoom Image
Scheme 2 Reductive deuteration of aromatic or unsaturated amines with deuterated ammonium formate. a Isolated overall yield of deuterated products [Dx ], chemical purity >95% (determined by NMR, HPLC or LC–MS). b Determined by MS analysis of a single experiment after filtration of the reaction mixture; replications of the experiments yielded slightly different mass distributions; [%D] value shows the relative distribution of mass units in the product.
Zoom Image
Figure 2 Examples of deuterated nitrosamine MS standards for LC–MS impurity quantification
Zoom Image
Scheme 3 HIE reactions of Kerr catalyst with tetrazoles.
Zoom Image
Figure 3 Reaction set-up and analytics of deuterium and tritium reactions
Zoom Image
Figure 4 Graphical overview in which temperature window the iridium catalysts are efficiently working (threshold 25%D) for HIE reactions with acetamidoacetophenone; the yellow star is indicating the highest deuterium incorporation
Zoom Image
Scheme 4 HIE reaction of amino acids and peptides with Kerr catalyst; 31 Ci/mmol = 1.15 TBq/mmol.
Zoom Image
Scheme 5 Deuteration of a variety of heterocycle-substituted tertiary amines with the Kerr catalyst
Zoom Image
Figure 5 Improved radiochemical yield for tritium reactions. Literature analysis over the last 20 years; 20 Ci/mmol = 0.74 TBq/mmol.
Zoom Image
Scheme 6 Hydrogen isotope exchange by homogeneous iridium(I) catalysis in aqueous buffers with deuterium or tritium gas
Zoom Image
Scheme 7 Deuteration with Tamm’s catalyst on anilines
Zoom Image
Scheme 8 Calculation of transition state and further understanding of Tamm’s catalyst behavior (14 Ci/mmol = 518 GBq/mmol)
Zoom Image
Figure 6 Deuteration with Burgess catalyst (2.3 Ci = 85 GBq; 9 Ci/mmol = 333 GBq/mmol).
Zoom Image
Scheme 9 Ruthenium-catalyzed HIE reaction of benzoic acid derivatives (5.5 Ci/mmol = 0.2 TBq/mmol)
Zoom Image
Scheme 10 Nanoparticle formation and heterogeneous HIE catalysis of anilines; IrNP = iridium nanoparticles
Zoom Image
Scheme 11 Ir nanoparticle assembly using D2 or T2
Zoom Image
Scheme 12 Photochemical deuteration and tritiation of peptides; photocatalyst 2 = Ir[dF(Me)], BTMG = Barton’s base
Zoom Image
Scheme 13 Photochemical deuteration and tritiation using D2O and T2O; 3DPA2FBN = 2,4,6-tris(diphenylamino)-3,5-difluorobenzonitrile
Zoom Image
Scheme 14 Combined photo- and metal-catalyzed deuteration of complex molecules, NHP1 = Ir2Cl2(COD)2, PC1= 4CzIPN (1,2,3,5-Tetrakis(carbazol-9-yl)-4,6-dicyanobenzene, 2,4,5,6-tetrakis(9H-carbazol-9-yl) isophthalonitrile)
Zoom Image
Scheme 15 Radical-mediated deuteration and tritiation of complex molecules.
Zoom Image
Scheme 16 Synthesis of tritiated taxane azide
Zoom Image
Figure 7 SGLT inhibitors AVE2268 and AVE8887
Zoom Image
Scheme 17 14C-labeling route to SGLT inhibitor AVE2268
Zoom Image
Scheme 18 14C-labeling route of AVE0991
Zoom Image
Scheme 19 Catalytic cycle of the carbon isotope exchange (CIE) of amino acids; RCY: radiochemical yield, red/blue: isotope incorporation in %, yield in brackets. a TE: trapping efficiency.
Zoom Image
Figure 8 Structure of 19F/18F-TAK875
Zoom Image
Scheme 20 Late-stage functionalization of peptides: iodination of tyrosines (top); benzylic C–C coupling of tyrosines or phenolic ethers (bottom)
Zoom Image
Scheme 21 Site-selective C–H functionalization of N-(hetero)aryl piperidines, morpholines, and piperazines.