**4. Lipid spin probes (oxazolidinyl or doxyl, proxyl)**

The development of spin labeling and spin probes expanded to lipids and membranes. In order to probe these biological structures, one needs a label that mimics or looks like a lipid and can be incorporated into a phospholipid membrane structure. As late as the late 1960s, one could only prepare an ester of a fatty acid with one of piperidine or pyrrolidine nitroxides, but one could not incorporate a probe somewhere in the middle of the lipid chain in order to probe various depths of a membrane. It was not until John Keana demonstrated that one can incorporate an oxazolidine ring at specifically placed ketone (keto) groups in a lipid, resulting in a rigid five membered ring fused to the lipid chain that was easily oxidized to the radical nitroxide (doxyl) [11]. This virtually led to a revolution in our ability to probe membrane structure and dynamics with structural and dynamic accuracy. Several of these compounds are shown below. It took a while before these were commercially available, however, the synthesis was reasonably straightforward and scientists in the area were willing to share their compounds with one another. This was a clear departure from the relatively straightforward chemistry of the piperidine, pyrroline and pyrrolidine based aminoxyl radicals that had been developed by the Russian groups up to that point. The synthesis is relatively straightforward: take a lipid of interest, which can be purchased as a halo-derivative or occasionally as the desired keto derivative. Then the oxazolidine ring is formed at this position on the chain, then oxidized to the radical. Spin labeled lipid probes became available with aminoxyl radical group at the 5-, 12-, or 16 position in the lipid chain, and later at other positions. The resulting biochemistry, i.e., to incorporate these lipid nitroxides at either the 2- or 3- position of a phospholipids, was fairly straightforward as the fatty acid interchange or ester interchange chemistry was already well known. The synthetic schemed and some example probes are shown below, with a phospholipid analog in Figure 6.

Some years later, the problem of the oxazolidine ring being essentially reversible, (i.e. hydrolyzable), a newer development involved the incorporation directly of a fivemembered ring (proxyl) into the structure of a lipid molecule at a strategically placed double bond. The chemistry again was somewhat sophisticated but straightforward; the synthetic route (Figure 7) leads to a side-chain-substituted 2,2,5,5-tetramethylpyrrolidine-Noxy1 (proxyl) nitroxide lipid spin probes from a commercially available nitrone is treated with an organometallic reagent, which after Cu2+-air oxidation gives a new intermediate nitrone, followed by a second selected organometallic reagent, which after Cu2+-air oxidation yields the proxyl spin probe [11]. The advantage of proxyl chemistry over oxazolidine chemistry in order to make lipid spin probes was that one could tailor the orientation of the N-O group with respect to the lipid axis. This became important since the hyperfine coupling constant along the z-axis of the label (i.e. directly above and perpendicular to the N-O plane) yielded a large splitting, upwards of 32G, that allowed a quite accurate estimate of the orientation, order parameter and dynamics of this portion of the lipid spin probe within the membrane.

8 Nitroxides – Theory, Experiment and Applications

phospholipid analog in Figure 6.

spin labeled analogs could be determined from the electron-electron dipole interaction. This was the first example of distance measurements involving two spin labels within a protein structure and, due to the fortuitous situation of a perfectly, rigidly bound spin label, distances could be determined precisely [10]. This study still remains the gold standard of

The development of spin labeling and spin probes expanded to lipids and membranes. In order to probe these biological structures, one needs a label that mimics or looks like a lipid and can be incorporated into a phospholipid membrane structure. As late as the late 1960s, one could only prepare an ester of a fatty acid with one of piperidine or pyrrolidine nitroxides, but one could not incorporate a probe somewhere in the middle of the lipid chain in order to probe various depths of a membrane. It was not until John Keana demonstrated that one can incorporate an oxazolidine ring at specifically placed ketone (keto) groups in a lipid, resulting in a rigid five membered ring fused to the lipid chain that was easily oxidized to the radical nitroxide (doxyl) [11]. This virtually led to a revolution in our ability to probe membrane structure and dynamics with structural and dynamic accuracy. Several of these compounds are shown below. It took a while before these were commercially available, however, the synthesis was reasonably straightforward and scientists in the area were willing to share their compounds with one another. This was a clear departure from the relatively straightforward chemistry of the piperidine, pyrroline and pyrrolidine based aminoxyl radicals that had been developed by the Russian groups up to that point. The synthesis is relatively straightforward: take a lipid of interest, which can be purchased as a halo-derivative or occasionally as the desired keto derivative. Then the oxazolidine ring is formed at this position on the chain, then oxidized to the radical. Spin labeled lipid probes became available with aminoxyl radical group at the 5-, 12-, or 16 position in the lipid chain, and later at other positions. The resulting biochemistry, i.e., to incorporate these lipid nitroxides at either the 2- or 3- position of a phospholipids, was fairly straightforward as the fatty acid interchange or ester interchange chemistry was already well known. The synthetic schemed and some example probes are shown below, with a

Some years later, the problem of the oxazolidine ring being essentially reversible, (i.e. hydrolyzable), a newer development involved the incorporation directly of a fivemembered ring (proxyl) into the structure of a lipid molecule at a strategically placed double bond. The chemistry again was somewhat sophisticated but straightforward; the synthetic route (Figure 7) leads to a side-chain-substituted 2,2,5,5-tetramethylpyrrolidine-Noxy1 (proxyl) nitroxide lipid spin probes from a commercially available nitrone is treated with an organometallic reagent, which after Cu2+-air oxidation gives a new intermediate nitrone, followed by a second selected organometallic reagent, which after Cu2+-air oxidation yields the proxyl spin probe [11]. The advantage of proxyl chemistry over oxazolidine chemistry in order to make lipid spin probes was that one could tailor the orientation of the N-O group with respect to the lipid axis. This became important since the

distance measurements by electron-electron dipolar interactions.

**4. Lipid spin probes (oxazolidinyl or doxyl, proxyl)** 

**Figure 6.** 16:0-7 Doxyl PC 1-palmitoyl-2-stearoyl-(7-doxyl)-sn-glycero-3-phosphocholine.

Even more rigid lipid probes were possible with the advent of racemic azethoxyl lipid probes nitroxides (called minimum steric perturbation spin labels). In the azethoxyl the nitrogen atom is actually embedded in the hydrocarbon chain. Cis-trans isomerism is possible and modeling suggests that the trans isomer should resemble a saturated lipid, whereas the cis isomer introduces a bend in the chain which approximates that observed with a cis carbon-carbon double bond.

The general synthetic route to the azethoxyl nitroxide spin labels is similar to that of the proxyl nitroxides, except that a different nitrone is used in the beginning (Figure8), where in this specific example, the trans isomer predominates [11].

Synthetic development was also carried out by several chemists in Ljubljana, Slovenia, as well as other synthetic organic labs, all of which were principally in Eastern Europe. In the U.S.A., the plight of an organic chemist attempting to obtain tenure in an academic department required the synthesis of complex natural products for the development of new synthetic reactions. Frequently the synthetic procedures for preparing these aminoxyl radicals, spin labels or spin probes were albeit modern but not new and novel; the organic chemist simply adapted the new, clever synthetic procedures to obtain the required label. It

wasn't until the late 1990s, or perhaps the new millennium, where chemistry departments accepted applied chemistry as a valid academic area of new ideas and novel techniques. Certainly, it was the synthetic organic chemist who solved this problem and, for that matter, most biophysical studies involving probes depend on clever synthetic abilities. EPR had a great advantage in membrane and cell studies and cell membranes since the technique did not require optical transparency, did not have the magnetic susceptibility problems encountered in NMR, and required a fairly low level of spin label doping of the biological system in order to obtain a strong, highly sensitive spectrum. Indeed, it is fair to say that EPR added a tremendous amount of knowledge to our understanding of lipid, membrane and related polymeric systems, which was a great complement to that learned from NMR, solid-state NMR and microscopic methods The real leadership in the implications of these problems started, again, in the McConnell lab at Stanford University and with people like Joe Seelig, Wayne Hubbell and others who followed. Nobel laureate Roger Kornberg was also a graduate student in this laboratory, and his work also was involved in studies of lipids and membranes through the use of spin labels and spin probes [12].

**Figure 7.** Synthesis of a proxyl nitroxide

### **5. Nucleic acid analogues**

The Bobst laboratory at the University of Cincinnati synthesized some very novel nucleotide analogues where the label was covalently tethered to various purine and pyrimidine rings in such a manner that the tether did not distort the DNA structure and was rigid enough to not create ambiguities in an interpretation of the backbone or sidechain motion of the polynucleotide where the label was incorporated [13]. A series of these novel, unique structures are shown in Figure 9. This work was then copied and extended by other groups, particularly the Seattle group (University of Washington) that also designed nucleotide analogues for probing DNA [14]. In all cases the syntheses were truly challenging, could only be carried out by very proficient organic chemists, and support the view of this author that synthetic organic chemistry is the rate-limiting step in many of these biophysical probe experiments.

**Figure 8.** Synthesis of an azethoxyl nitroxide

through the use of spin labels and spin probes [12].

**Figure 7.** Synthesis of a proxyl nitroxide

**5. Nucleic acid analogues** 

experiments.

wasn't until the late 1990s, or perhaps the new millennium, where chemistry departments accepted applied chemistry as a valid academic area of new ideas and novel techniques. Certainly, it was the synthetic organic chemist who solved this problem and, for that matter, most biophysical studies involving probes depend on clever synthetic abilities. EPR had a great advantage in membrane and cell studies and cell membranes since the technique did not require optical transparency, did not have the magnetic susceptibility problems encountered in NMR, and required a fairly low level of spin label doping of the biological system in order to obtain a strong, highly sensitive spectrum. Indeed, it is fair to say that EPR added a tremendous amount of knowledge to our understanding of lipid, membrane and related polymeric systems, which was a great complement to that learned from NMR, solid-state NMR and microscopic methods The real leadership in the implications of these problems started, again, in the McConnell lab at Stanford University and with people like Joe Seelig, Wayne Hubbell and others who followed. Nobel laureate Roger Kornberg was also a graduate student in this laboratory, and his work also was involved in studies of lipids and membranes

The Bobst laboratory at the University of Cincinnati synthesized some very novel nucleotide analogues where the label was covalently tethered to various purine and pyrimidine rings in such a manner that the tether did not distort the DNA structure and was rigid enough to not create ambiguities in an interpretation of the backbone or sidechain motion of the polynucleotide where the label was incorporated [13]. A series of these novel, unique structures are shown in Figure 9. This work was then copied and extended by other groups, particularly the Seattle group (University of Washington) that also designed nucleotide analogues for probing DNA [14]. In all cases the syntheses were truly challenging, could only be carried out by very proficient organic chemists, and support the view of this author that synthetic organic chemistry is the rate-limiting step in many of these biophysical probe
