Hey Geisha, Hold on !

Figure 1. Taken from Ref. [1].

The Angew. Chem. paper I am sharing with you this time is impressive, it is on the chemistry-biology interface. [1] It is about bio-conjugation , which is applying a chemical reaction to a biological macromolecule. Let’s deal with both fields in detail, one by one.

Bioconjugation is useful, because we can start to attach unnatural or unconventional chemical substrates onto biological molecules, such as proteins, complex carbohydrate architectures or nucleic acid-based compounds. While these methodologies have found great uses in synthetic biology, the design of the reactions is often challenging. Unlike conventional organic substrates, the reactants we are dealing with here are biological molecules, which are sensitive to high temperature, extreme pH, and possess loads of stereogenic centers. Conventional wisdom shows us that we can lower the activation energy barrier of a reaction (i.e. increasing the reaction rate) by increasing the reaction temperature. However, this does not help in the case of biological molecules, as the high temperature, or a sudden perturbation in pH, can lead to negative results, such as the denaturation of proteins or possible racemizations. So, the best methods for bioconjugation should work optimally at physiological temperature (37^oC) , narrow pH range, and also work well in an aqueous media.

How can we incorporate unnatural chemical functionalities into a supposedly natural biomolecule? Molecular biology methods, of course! In this case, a protein was expressed with an unnatural amino acid (UAA), with a carbon-carbon triple bond (alkyne) on it. The alkyne can act as a ‘handle’ for other reactions, and in the past, click chemistry and Sonogashira coupling has already been successfully employed.

Figure 2. The Glaser-Hay coupling reaction.

Figure 3. The catalytic cycle of Glaser-Hay coupling reaction.

The reaction in focus is known as the Glaser-Hay coupling reaction (Figure 2). It is a copper-catalyzed coupling reaction between 2 carbon-carbon triple bonds. Previous studies have concluded that the reaction probably involved a dimeric copper acetylide intermediate, and the mechanism involved an oxidation of the copper species, from +1  to +2 and then +3, and finally back to +1, continuing the catalytic cycle (Figure 3). Traditionally, this was a homo-coupling reaction, which meant you had 2 identical coupling partners to get the dimeric, symmetrical product. This unfortunately hinders its potential applications. Because what we want is a CROSS-coupling (hetero-coupling) reaction, in which you have 2 different coupling partners, A and B, to form an A-B type product (rather than an A-A or B-B type product). How about adding 2 different substrates to the Glaser-Hay system? Nope, it does not work well, as we almost alyways have a statistical amount of homo-coupling products, and that hinders purifications too. There is good news, though, as recently novel systems have been developed to lead predominately to hetero-coupled partners. For example, Lei and Agrofoglio independently developed co-catalyzed system involving a Ni (II) / Cu (I) system, and these methods work well for a variety of alkynic substrates. On the other hand, if one of the coupling partners is attached to a solid support, heterocouplings also predominate. Glaser-Hay reactions work well in organic solvents, yet it is rather unprecedented to work in an aqueous media. This article shows us that possibility.

Figure 4. The substrate for the coupling reaction. Taken from Ref. [1].

The researchers started with some model reactions on small molecules before embarking on biomolecules (Figure 4, a). With phenylacetylene and propargyl alcohol respectively, they could achieve respectively homo-coupled products in an aqueous solution with CuI / TMEDA as the catalytic system. They had to premix these 2 reagents to establish the copper complex before it could be bound to an alkyne substrate and start catalysis (Figure 5).

 Figure 5. The first stage of the catalytic cycle. Taken from Ref. [1].

To achieve unconventional biological products for their methodolgy, the researchers installed an unconventional amino acid, ‘propargyl phenylalanine’, at position 151 of the green fluorescent protein substrates they used, which was known as pPrF-GFP (Figure 4, b). They coupled this substrate with ‘AlexaFluor 488 modified alkyne’, and succeeded in making the hetero-coupled product (Figure 6, 7). They have also explored the substrate scope of their method. For solid-supported reactions, they have achieved similar yields with a propargyl or a hexyl-derived alkyne, and also in other types of biologically-related substrates (Figure 8). The results were easily visualized, as a successful coupling reaction would lead to an observable fluorescent signal, or by SDS-PAGE analysis (Figure 7).  

Figure 6. Glaser-Hay bioconjugation. Taken from Ref. [1].

Figure 7. The results of the Glaser-Hay bioconjugations as illustrated by a normal SDS-PAGE gel and a fluorescent version. In diagram A, we can see that the unconventional amino acid, pPrF, has been successfully incorporated into the GFP (Lane 2), because there is no GFP spot in Lane 3, when pPrF is absent. In diagram B, we can observe that the coupling reaction has been successful carried out because in Lane 2 and 4, which corresponds to 2 different temperatures, and with both copper catalyst and pPrF-GFP present, this substrate is coupled to the fluorophore ‘Alexa’. Therefore, fluorescene signals are evident in those lanes, signifying successful C-C bond formation. This is not the case for the other control experiments (Lane 3, 5, 6, 7), where one or both of the aforementioned components are absent. When the researches use a wild-type GFP – that means one that has no pPrF incorporated, of course no coupling occurs.

Figure 8. Other substrates. Taken from Ref. [1].

The researchers have discovered a possible unproductive side reaction - a Cu (I) -catalyzed oxidative degradation. Another interesting observation was that the reaction rates were comparable at a low (4 ^oC) and high temperature (37 ^oC). They attributed this to a  protein degradation at the higher temperature, inhibiting reaction progress.

Some personal opinions

This paper is absolutely smashing, and I hope there will be further developments on the project. In the paper, the researchers have employed 2 different alkynic substrates, a propargyl and a hexyl one. I wonder if there will be any different, in terms of reaction rates, as the chemical environment next to the 2 respective triple bonds are surely distinctive – in the propargyl case, the triple bond and the oxygen is only separated by a CH₂ group, while in the hexyl alcohol case, the oxygen and the triple bond were separated by a much longer spacer. A further question would be, can we use a more hindered propargyl-type secondary ether, as we may even generate a possible stereogenic center close to the biological site (Figure 9)? The ‘double conjugated alkyne’ should serve as a nice handle for further reactions to occur – how about a thiol-yne coupling or a nucleophilic addition of ROH onto the terminal alkyne?

Figure 9. A potential asymmetric Glaser-Hay coupling reaction?

The second question - is it possible to use a Cu (II) salt, such as CuSO₄ , instead to effect the coupling reaction? Obviously, this would not have the same mechanistic similarity like the Glaser-Hay type reaction, yet homocoupling involving Cu (II) salts do exist, e.g. Cu(OAc)₂ for Eglinton reaction. If this is OK, then it will increase the versatility of the reaction as the Cu (II) salts are readily accessible.

It would be really exciting if this research topic can be developed much further!

by Ed Law

28/6/2015

Reference:

1. Development and Optimization of Glaser–Hay Bioconjugations

J. S. Lampkowski, J. K. Villa, T. S. Young, and D. D. Young

Angew. Chem. Int. Ed., 2015, asap.

DOI: 10.1002/anie.201502676

Growing out of Copper's Pants

Figure 1. Porphyry copper deposits. 

(Taken from http://pubs.usgs.gov/fs/fs053-03/fs053-03.html)

You see, I always have a strong interest in Geology and Geomorphology, and indeed I have been a tutor of the undergraduate Environmental Chemistry course for some time! I read ‘Nature Geoscience’ regularly and this month I have read a fascinating article about ‘porphyry copper deposit’. [1]

Figure 2. Formation of porphyry copper deposits. Taken from Source [2].

The 'golden child' here is copper (should be called 'copper child?), and these copper deposits indeed have high economical values. These deposits are porphyritic, and intrusive (hence plutonic) in nature. The positions of these deposits are often found near the convergent plate margins, and above the subducting plate. The magmatic (hydrothermal) fluid first rises through the continental crust. When the hydrothermal fluid rises, the rate of its cooling increases gradually, as it passes through the area of the fractures and intrusions. Mineralization leads to the formation of the porphyry copper deposits (Figure 2). [2]

A key theme of the article is about tectonic uplift, which is the 'lifting-up' of the surface of the mountain belts, in opposite to the direction of gravity. One of the components is known as 'exhumation', which is defined as 'the displacement of rocks with respect to surface'. Quantitatively, the rate of exhumation is related to the rate of erosion or the removal of overburden by tectonic processes. So, exhumation comes hand in hand with exogenic denudation processes. Erosion will wear away the mountains, and so the metamorphic rocks or reserves from below will become exposed at the surface.[3]

The Nature Geoscience article shows the relationship between climate and the effects it may cause to the orogenic geomorphology. The researches have studied Cenozoic porphyry copper deposits at convergent tectonic settings. From various data, they have discovered that a higher precipitation, which means a higher erosion rate by rain water, will lead to faster exhumation. The consequence of rapid exhumation is the sparsely distributed porphyry copper deposits, with a relatively younger age. By contrast, lower precipitation, which corresponds to an arid region, will lead to slower exhumation, and the copper deposits are more abundant and relatively aged. Therefore, in a sense the age of the copper deposits is related to the exhumation rates of the active orogens, and so this can serve as a 'sensor' to uncover the geological phenomenon. 

Upon the completion of this article, I have discovered that it also has been covered on the web already. Well, that means this research is really significant for many! [4]

By Ed Law

19/6/2015

  

Reference and sources:

1. A climate signal in exhumation patterns revealed by porphyry copper deposits

Brian J. Yanites and Stephen E. Kesler

Nature Geoscience 2015, 8, 462–465.

doi:10.1038/ngeo2429

2. http://ns.umich.edu/new/releases/22883-a-climate-signal-in-the-global-distribution-of-copper-deposits

3. Surface uplift, uplift of rocks, and exhumation of rocks.

P. England, P. Molnar

Geology, 1990, 1173-1177.

4. http://www.indicoresources.com/s/CopperPorphyryDeposits.asp

Forget-Me-Not, Iron Man

Figure 1. Taken from [1].

One aspect of organic chemistry that has always fascinated me is the possibility of ‘chemical memory’. In an organic molecule, the information is designated in the structure of the molecule itself – what formula (i.e. what atoms in it), how the atoms are arranged and most importantly, its stereochemistry (i.e. its 3D-arrangement). For the stereochemical information, chirality is the more important aspect, as it is the signature that distinguishes two molecules which has exactly the same atom arrangement and same formula.

It is well known that some chemical reactions can destroy the stereochemical information of the starting material. Take the classic example, a first-order, nucleophilic reaction – a S_N1 reaction (Figure 2). The mechanism dictates that the leaving group first departs to generate a planar, positive-charged carbocation. The essential detail here is that it is planar. Because of this particular shape, an incoming nucelophile will have a 50% / 50% chance of either attacking from the top of the carbocation, or from the bottom. That means, judging from the stereochemistry from the product (‘R’ or ‘S’ form), that is absolutely no way you will know which starting isomer makes the product, because the sterochemical information is lost upon the formation of the planar carbocation.

 Figure 2. For the S_N1 type reaction, the stereochemical information is lost upon the formation of the relatively stable tertiary / secondary carbocation, leading to a racemization of product. Taken from Clayden et. al., Organic Chemistry P.421.

In contrast, this is not the case for a S_N2 reaction, which always results in an inversion of stereochemistry (if there is no neighbouring group participations), which means, for example, if you have a product as a ‘S’-isomer, you know it originates from a starting reactant in ‘R’-form, and vice versa (Figure 2).

The loss of stereochemical information can be really tragic, especially in the field of asymmetric synthesis, as a racemization via a proposed strategy basically suggests that your method is heading towards a dead-end. There are, however, examples that stereochemical information can be preserved. One of them is the ferrocene-based carbocation, which is conformationally stable, and therefore when the carbocation is attacked by a nucelophile, it will lead to a retention of configuration – which means the molecule ‘remembers’ its past stereochemistry (Figure 3). This type of chemistry is good news – because by design, we can control the outcome of the reaction confidently now!

Figure 3. Taken from [1].

The Organic Letters article I share with you this time is related to molecular memory, and the reaction is the classic Friedel-Crafts reaction [1]. As you may have learnt in high school, the key step of a Friedel-Crafts alkylation is the generation of a stable carbocation, via the action of a Lewis Acid (AlCl₃, FeBr₃, to name a few). The researchers have demonstrated that, with the inclusion of a silicon functionality, a retention of configuration can be achieved for the product, that means the compound has shown ‘molecular memory’ and remembers its initial stereochemical configuration.

The model reaction of the substrates without a silicon group shows that, upon reaction, a racemic mixture results. So, the planar carbocation leads to same amount of both the isomers (Figure 4).  

Figure 4. Control experiment leads to racemization, as expected. Taken from [1].

Upon the use of the silyl group and an iron salt as Lewis acid, a retention results for the product. Note the alcohol group in the starting material and the indole ring in the product are both pointing into the plane (Figure 5).

Figure 5. Retention of configuration from silyl substrates. Taken from [1].

They have provided a mechanistic rationale. They believe that the ‘molecular memory’ originated from the β –silyl effect, which leads to the stabilization of the carbacationic intermediate. Thus, the iron salt activates the –OH group and generates the conformationally stable carbacation, then the indole attacks and leads to the final product, with a net retention of configuration (Figure 6).

 Figure 6. Mechanistic rationale of the Friedel-Crafts alkylation. Taken from [1].

With iron man, I can remember my past now!

 by Ed Law

15/6/2015

The β‑Silyl Effect on the Memory of Chirality in Friedel−Crafts Alkylation Using Chiral α‑Aryl Alcohols

Toshiki Nokami,Yu Yamane, Shunsuke Oshitani, Jun-ka Kobayashi, Shin-ichiro Matsui, Takashi Nishihara, Hidemitsu Uno, Shuichi Hayase, and Toshiyuki Itoh

Org. Lett., 2015, asap

DOI: 10.1021/acs.orglett.5b01582

Triangle goes viral

Figure 1. Taken from [1].

Just a quick one here. This is a paper from Organic Letters, where the researchers have made some fluorinated analogues of an inhibitor against Hepatitis C virus. An interesting aspect is that the chemical structure contains a triangle – no, I mean cyclopropyl, the 3-membered carbon ring.

Recently, it becomes known that a major strategy against Hepatitis C virus was to target a protease (NS3/4A), which is involved in the replication of the virus. A known inhibitor of this enzyme is the compound 2 in Figure 1. Not only this peptidomimetic have a di-peptide backbone, it also consists of a cyclopropyl-amino acid functionality. That is why if we want to start making analogues resembling this compound, we have to start with the cyclopropyl core.

The researchers of this paper decided to go one step further - they want to test whether the inclusion of a fluorine atom, bonding directly to the carbon atom in the cyclopropyl core, would lead to any improvement of the inhibitor.

The reason why this paper caught my attention was because I was fascinated by the cyclopropyl type structure, an also its synthesis, nevertheless we will not miss any other details.

Figure 2. Synthesis of the fluorinated cyclopropyl amino ester building block. Taken from [1].

The first stage is to make the protected, fluorinated cyclopropyl amino ester 8 (Figure 2). Using ethyl dibromofluoroacetate, they carried out a cycloproponation with the aminoacrylate 7, with Zn/LiCl at low temperature, with dropwise addition. Indeed, LiCl can accelerate many organozinc and also organomagnesium (Grignard reaction), but one thing important about LiCl (which I can convince you because I have done some related experiments). LiCl is really hygroscopic, so you have to heat it up and dry it under vacuum before use. Except this precaution, LiCl really helps to promote the reaction, and literature abounds with its use.  The resulting cyclopropyl amino ester was stable to column chromatography, and they got that with a reasonably great yield. Their next key challenge was to install the exocyclic double bond, sort of conjugated to the cyclopropyl ring. That involved a series of steps, and the pen-ultimate step involved a Wittig reaction to put in the double bond. After an acidic hydrolysis, they get the amino ester hydrochloride salt 6. 

Figure 3. Completion of Synthesis. Taken from [1].

The reason why they made the compound 6 was because they wanted to develop a strategy to make a fluorinated version of Simeprevir, and compound 6 was actually one of the 4 building blocks they are going to put together at the end. Indeed, their synthesis indeed exposed some of the chemical properties of the building blocks, including compound 6, from the side-reactions they encountered throughout the optimization (Figure 3). For example, a relative higher temperature led to the ring-opening of the cyclopropyl, and indeed they can monitor this because of the distinct 19F NMR shifts of the fluorine atoms in the decomposition products and the cyclopropyl fluorine (Figure 4). They counteracted the problem by lowering the temperature to -15 Celsius. The other key reactions to join the fragments together included a Mitsunobu, a ring-closing metathesis and a mixed anhydride coupling reaction. So, they have devised a novel strategy towards fluorinated analogues and they have also submitted their compounds to some preliminary antiviral activities studies.

Figure 4. 19F NMR showed that the chemical shift of the cyclopropyl fluorine should be very different from that of its decomposition products, which originated from a ring-opened intermediate. Indeed, it would be interesting if this olefinic intermeidate could be trapped by a quenching experiment, or some in-situ NMR experiments could be carried out to study the evolution of this reaction. Taken from Ref. [1].

by Ed Law

12/6/2015

Reference:

1. Toward the Synthesis of Fluorinated Analogues of HCV NS3/4A Serine Protease Inhibitors Using Methyl α-Amino-β-fluoro-β-vinylcyclopropanecarboxylate as Key Intermediate

G. Milanole, F. Andriessen, G. Lemonnier, M. Sebban, G. Coadou, S. Couve-Bonnaire, J.-F. Bonfanti, P. Jubault, and X. Pannecoucke

Org. Lett., 2015, asap

DOI: 10.1021/acs.orglett.5b01216