Photoactivatable Reagents, Including Photoreactive Crosslinkers and Caged Probes—Section 5.3

Page Contents

• Nonfluorescent Photoreactive Crosslinking Reagents
• Other Photoreactive Reagents
• Caged Probes and Their Photolysis
• Caged Fluorescent Dyes
• Caging Carboxylic Acids with DMNPE
• Data Table
• Ordering Information

This section describes two types of photoactivatable probes: products that form short-lived, high-energy intermediates that can chemically couple to nearby residues, and "caged" probes that are designed to be biologically inactive until UV light–mediated photolysis releases a natural product. Photolysis of each of these photoactivatable probes can be accomplished with high spatial and temporal resolution, releasing active probe at the site of interest.

In contrast to chemical crosslinking reagents (Chemical Crosslinking Reagents—Section 5.2), which are often used to prepare bioconjugates, photoreactive crosslinking reagents are important tools for determining the proximity of two sites. These probes can be used to define relationships between two reactive groups that are on a single protein, on a ligand and its receptor, or on separate biomolecules within an assembly. In the latter case, photoreactive crosslinking reagents can potentially reveal interactions among proteins, nucleic acids and membranes in live cells. The general scheme for defining spatial relationships usually involves photoreactive crosslinking reagents that contain a chemically reactive group as well as a photoreactive group. These crosslinkers are first chemically reacted with one molecule, for example a receptor ligand, and then this modified molecule is coupled to a second molecule, for example the ligand's receptor, using UV illumination. Depending on the reactive properties of the chemical and photoreactive groups, these crosslinkers can be used to couple like or unlike functional groups.

We offer three types of photoreactive reagents for covalent labeling:

• Simple aryl azides that upon illumination (usually at <360 nm) generate reactive intermediates that form bonds with nucleophilic groups (Figure 5.3.1)
• Fluorinated aryl azides that upon UV photolysis generate reactive nitrenes, thereby producing more C–H insertion products than the simple aryl azides (Figure 5.3.2)
• Benzophenone derivatives that can be repeatedly excited at <360 nm until they generate covalent adducts, without loss of reactivity (Figure 5.3.3)

Figure 5.3.1 Photoreactive crosslinking reaction of a simple aryl azide.

Figure 5.3.2 Photoreactive crosslinking reaction of a fluorinated aryl azide.

Figure 5.3.3 Photoreactive crosslinking reaction of a benzophenone derivative.

Simple Aryl Azide Crosslinker

The "transferable" aryl azide N-((2-pyridyldithio)ethyl)-4-azidosalicylamide (PEAS; AET) is a unique reagent for assessing protein–protein or protein–nucleic acid interactions. This aryl azide undergoes disulfide–thiol interchange of its pyridyldisulfide groups with the thiol groups of biomolecules to form mixed disulfides in the same way as SPDP (S1531, Chemical Crosslinking Reagents—Section 5.2). UV photolysis induces covalent crosslinking to residues or biomolecules adjacent to the crosslinker. The mixed disulfide can then be cleaved with DTT or TCEP (D1532, T2556; Chemical Crosslinking Reagents—Section 5.2). If the phenolic PEAS reagent is radioiodinated before the coupling and photolysis steps, then only the resulting target biomolecule will be radioactive at the conclusion of the reaction.

Fluorinated Aryl Azides: True Nitrene-Generating Reagents

Although the simple aryl azides may be initially photolyzed to electron-deficient aryl nitrenes, it has been shown that these rapidly ring-expand to form dehydroazepines—molecules that tend to react with nucleophiles rather than form C–H insertion products. In contrast, Keana and Cai have shown that the photolysis products of the fluorinated aryl azides are clearly aryl nitrenes and undergo characteristic nitrene reactions such as C–H bond insertion with high efficiency. Moreover, conjugates prepared from the amine-reactive succinimidyl ester of  4-azido-2,3,5,6-tetrafluorobenzoic acid (ATFB, SE) may have quantum yields for formation of photocrosslinked products that are superior to those of the nonfluorinated aryl azides. An important application of the succinimidyl ester of ATFB is the photofunctionalization of polymer surfaces (Figure 5.3.4).

Figure 5.3.4 Schematic showing attachment of an amine-modified oligonucleotide to a surface using the photoreactive crosslinking reagent 4-azido-2,3,5,6-tetrafluorobenzoic acid, succinimidyl ester (ATFB, SE).

Benzophenone-Based Photoreactive Reagents

Benzophenones generally have higher crosslinking yields than the aryl azide photoreactive reagents. Benzophenone maleimide has been used for efficient irreversible protein crosslinking of actin, calmodulin, myosin, tropomyosin, troponin, ATP synthase and other proteins. The succinimidyl ester of 4-benzoylbenzoic acid and benzophenone isothiocyanate have proven useful for synthesizing photoreactive peptides and oligonucleotides. A benzophenone-labeled ATP probe (BzBzATP) is described in Other Photoreactive Reagents below.

Ethidium Monoazide for Photoreactive Fluorescent Labeling of Nucleic Acids

Ethidium monoazide (E1374) can be photolyzed in the presence of DNA or RNA to yield fluorescently labeled nucleic acids, both in solution and in cells. The efficiency of the irreversible photolytic coupling of ethidium monoazide, which intercalates into nucleic acids like ethidium bromide, is unusually high (>40%). The membrane-impermeant ethidium monoazide is reported to label only those cells with compromised membranes and can therefore serve as a fixable cell viability probe. This property, allied to the blocking of transcription caused by photoreaction of ethidium monoazide with DNA, provides a method for suppressing PCR amplification of dead-cell DNA. Similarly, multiphoton-targeted photochemistry of vertebrate cells labeled with ethidium monoazide was used to selectively inactivate gene expression. A mixed population of live and dead cells labeled with ethidium monoazide retains its staining pattern after aldehyde-based fixation, thereby reducing the investigator's exposure to potentially pathogenic cells during cell viability analysis.

Bimane Azide for Photoaffinity Labeling of Proteins

Bimane azide is a small blue-fluorescent photoreactive alkyl azide (excitation/emission maxima ~375/458 nm) for photoaffinity labeling of proteins. This reactive fluorophore's small size may reduce the likelihood that the label will interfere with the function of the biomolecule, an important advantage for site-selective probes.

Photoreactive ATP Derivative for Labeling Nucleotide-Binding Proteins

Functional ion channels can be assembled from both homomeric and heteromeric combinations of the seven P2X purinergic receptor subunits so far identified (P2X_1–7). Due to the lack of specific agonists or antagonists for P2X receptors, it is difficult to determine which receptor subtypes mediate particular cellular responses. BzBzATP (2'-(or 3'-)O-(4-benzoylbenzoyl)adenosine 5'-triphosphate) is one of the most potent and widely used P2X receptor agonists, BzBzATP also has more general applications for site-directed irreversible modification of nucleotide-binding proteins via photoaffinity labeling.

Flash photolysis of photoactivatable or "caged" probes provides a means of controlling the release—both spatially and temporally—of biologically active products or other reagents of interest. The chemical caging process may also confer membrane permeability on the caged ligand, as is the case for caged cAMP and caged luciferin. Our extensive selection of caged nucleotides, second messengers (), chelators and neurotransmitters has tremendous potential for use with both live cells and isolated proteins. These caged probes provide researchers with important tools for delivering physiological stimuli by naturally active biomolecules with spatial and temporal precision that far exceeds that of microinjection or perfusion. A recent review by Ellis-Davies describes the optical and chemical properties of many of our caged compounds, as well as of several common caging groups.

Caging Groups

The caging moiety (Properties of six different caging groups—Table 5.2) is designed to maximally interfere with the binding or activity of the molecule. It is detached in microseconds to milliseconds by flash photolysis at ≤360 nm, resulting in a pulse (concentration jump) of active product. Uncaging can easily be accomplished with UV illumination in a fluorescence microscope or with a UV laser or UV flashlamp. Low-cost light-emitting diodes (LED) and 405 nm violet diode lasers are providing increased access to experimentation using caged compounds. The effects of photolytic release are frequently monitored either with fluorescent probes that measure calcium, pH, other ions or membrane potential, or with electrophysiological techniques.

Most of the caged reagents described in the literature have been derivatives of o-nitrobenzylic compounds. The nitrobenzyl group is synthetically incorporated into the biologically active molecule by linkage to a heteroatom (usually O, S or N) as an ether, thioether, ester (including phosphate or thiophosphate esters), amine or similar functional group. Both the structure of the nitrobenzylic compound and the atom to which it is attached affect the efficiency and wavelength required for uncaging. We currently use six different photolabile protecting groups in our caged probes. Their properties are summarized in Properties of six different caging groups—Table 5.2.

• Probes caged with our α-carboxy-2-nitrobenzyl (CNB) caging group generally have the most advantageous properties. These include good water solubility, very fast uncaging rates in the microsecond range, high photolysis quantum yields (from 0.2–0.4) and biologically inert photolytic by-products. Although the absorption maximum of the CNB caging group is near 260 nm, its absorption spectrum tails out to approximately 360 nm, allowing successful photolysis using light with wavelengths ≤360 nm.
• The 1-(2-nitrophenyl)ethyl (NPE) caging group has properties similar to those of CNB and can also be photolyzed at ≤360 nm.
• As compared with CNB and NPE, the 4,5-dimethoxy-2-nitrobenzyl (DMNB) and 1-(4,5-dimethoxy-2-nitrophenyl)ethyl (DMNPE) caging groups have longer-wavelength absorption (absorption maximum ~355 nm) and therefore absorb 340–360 nm light more efficiently. However, photolysis rates and quantum yields of DMNB- and DMNPE-caged probes are generally lower than those obtained for CNB-caged probes.
• The 5-carboxymethoxy-2-nitrobenzyl (CMNB) caging group provides an absorption maximum of intermediate wavelength (absorption maximum ~310 nm), while imparting significant water solubility to the caged probe. Its photolysis rate and quantum yield are intermediate between those of CNB- and DMNB-caged probes.
• The nitrophenyl (NP) caging group is available on the caged calcium reagent NP-EGTA (N6802), a photolabile Ca²⁺ chelator that can be used to rapidly deliver a pulse of Ca²⁺ upon illumination with ultraviolet light, with a high photolysis quantum yield of 0.23.

Experiments utilizing probes caged with any of the above caging groups, except the CNB caging group, may require the addition of dithiothreitol (DTT, D1532; Introduction to Thiol Modification and Detection—Section 2.1). This reducing reagent prevents the potentially cytotoxic reaction between amines and the 2-nitrosobenzoyl photolytic by-products.

Caged Nucleotides

Photoactivatable nucleotides and phosphates have contributed significantly to our understanding of cytoskeleton dynamics, signal transduction pathways and other critical cellular processes. Some caged nucleotides are available with a choice of caging group:

• Caged ATP (A1048), which has been shown to release ATP in skinned muscle fibers, sarcoplasmic reticulum vesicles, submitochondrial particles and membrane fragments containing Na⁺/K⁺-ATPase
• Caged ADP, which has been used to investigate the molecular basis of contraction of skeletal muscle fibers, as well as transport by an ADP/ATP carrier
• Caged cAMP, which is cell-permeant and rapidly photolyzed to cAMP
• Caged inositol 1,4,5-triphosphate and caged cADP-ribose, which are important probes for second messenger studies (Calcium Regulation—Section 17.2)

NPE-caged Ins 1,4,5-P₃ can be used to generate rapid and precisely controlled release of Ins 1,4,5-P₃ in intact cells () and is widely employed in studies of Ins 1,4,5-P₃–mediated second messenger pathways. NPE-caged Ins 1,4,5-P₃ is a mixture of the physiologically inert, singly esterified P⁴ and P⁵ esters and does not contain the somewhat physiologically active P¹ ester. NPE-caged Ins 1,4,5-P₃ exhibits essentially no biological activity prior to photolytic release of the biologically active Ins 1,4,5-P₃.

Cyclic ADP-ribose (cADP-ribose) is a potent intracellular Ca²⁺–mobilizing agent that functions as a second messenger in an Ins 1,4,5-P₃–independent pathway. NPE-caged cADP-ribose induces Ca²⁺ mobilization in sea urchin egg homogenates only after photolysis, and this Ca²⁺ release is inhibited by the specific cADP-ribose antagonist 8-amino-cADP-ribose. Furthermore, when microinjected into live sea urchin eggs, NPE-caged cADP-ribose was shown to mobilize Ca²⁺ and activate cortical exocytosis after illumination with a mercury-arc lamp.

Caged Ca²⁺ Reagents: NP-EGTA and DMNP-EDTA

Caged ions and caged chelators can be used to influence the ionic composition of both solutions and cells, particularly for ions such as Ca²⁺ that are present at low concentrations under normal physiological conditions. Developed by Ellis-Davies and Kaplan, nitrophenyl EGTA (NP-EGTA) is a photolabile Ca²⁺ chelator that exhibits a high selectivity for Ca²⁺ ions, a dramatic increase in its K_d for Ca²⁺ upon illumination (from 80 nM to 1 mM) and a high photolysis quantum yield (0.23). NP-EGTA's affinity for Ca²⁺decreases ~12,500-fold upon photolysis. Furthermore, its K_d for Mg²⁺ of 9 mM makes NP-EGTA essentially insensitive to physiological Mg²⁺ concentrations. We offer the tetrapotassium salt of NP-EGTA (N6802). The NP-EGTA salt can be complexed with Ca²⁺ to generate a caged Ca²⁺ reagent that will rapidly deliver Ca²⁺ upon photolysis (Figure 5.3.5). The cell-permeant AM ester of NP-EGTA does not bind Ca²⁺ unless its AM ester groups are removed. NP-EGTA AM can serve as a photolabile chelator in cells because, once converted to NP-EGTA by intracellular esterases, it will bind free Ca²⁺ until photolyzed with UV light.

The first caged Ca²⁺ reagent described by Kaplan and Ellis-Davies was 1-(4,5-dimethoxy-2-nitrophenyl) EDTA (DMNP-EDTA), which they named DM-Nitrophen (now a trademark of Calbiochem-Novabiochem Corp.). Because its structure more resembles that of EDTA than EGTA, we named it as a caged EDTA derivative (Figure 5.3.6). Upon illumination, DMNP-EDTA's affinity for Ca²⁺decreases ~600,000-fold and its K_d for Ca²⁺ rises from 5 nM to 3 mM. Thus, photolysis of DMNP-EDTA complexed with Ca²⁺ results in a pulse of free Ca²⁺. DMNP-EDTA has a stronger absorbance at longer wavelengths than does NP-EGTA (Figure 5.3.7), which facilitates uncaging. Furthermore, DMNP-EDTA has significantly higher affinity for Mg²⁺ (K_d = 2.5 µM) than does NP-EGTA (K_d = 9 mM), making it a potentially useful caged Mg²⁺ reagent. Two reviews by Ellis-Davies discuss the uses and limitations of DMNP-EDTA.

Figure 5.3.5 NP-EGTA (N6802) complexed with Ca2+. Upon illumination, this complex is cleaved to yield free Ca2+ and two iminodiacetic acid photoproducts. The affinity of the photoproducts for Ca2+ is ~12,500-fold lower than that of NP-EGTA.

Figure 5.3.6 DMNP-EDTA complexed with Ca2+. Upon illumination, this complex is cleaved to yield free Ca2+ and two iminodiacetic acid photoproducts. The affinity of the photoproducts for Ca2+ is ~600,000-fold lower than that of DMNP-EDTA.

Figure 5.3.7 Spectral comparison of equimolar concentrations of the caged Ca2+ reagents NP-EGTA (N6802, red line) and DMNP-EDTA (blue line), illustrating the optimal wavelengths for photolysis and subsequent release of Ca2+ from these chelators. Spectra were taken in 100 mM KCl and 30 mM MOPS buffer containing 39.8 µM free Ca2+ at pH 7.2.

Diazo-2: A Photoactivatable Ca²⁺ Knockdown Reagent

In contrast to NP-EGTA and DMNP-EDTA, diazo-2 is a photoactivatable Ca²⁺ scavenger. Diazo-2, which was introduced by Adams, Kao and Tsien, is a relatively weak chelator (K_d for Ca²⁺ = 2.2 µM). Following flash photolysis at ~360 nm, however, cytosolic free Ca²⁺ rapidly binds to the diazo-2 photolysis product, which has a high affinity for Ca²⁺ (K_d = 73 nM). Intracellular loading of NP-EGTA, DMNP-EDTA and diazo-2 is best accomplished by patch pipette infusion with the carboxylate salt form of the caged compound added to the internal pipette solution at 1–10 mM. These reagents are increasingly being applied in vivo for controlled intervention in calcium-regulated fundamental processes in neurobiology and developmental biology.

Caged Amino Acid Neurotransmitters

Once activated, caged amino acid neurotransmitters rapidly initiate neurotransmitter action (Figure 5.3.8), providing tools for kinetic studies of receptor binding or channel opening. Caged carbamylcholine (N-(CNB-caged) carbachol) and caged γ-aminobutyric acid (O-(CNB-caged) GABA), as well as two caged versions of L-glutamic acid, are all biologically inactive before photolysis.

Figure 5.3.8 CNB-caged L-glutamic acid. The CNB-caging group is rapidly photocleaved with UV light to release L-glutamic acid.

Caged Luciferin

Luciferase produces light by the ATP-dependent oxidation of luciferin. The 560 nm chemiluminescence from this reaction peaks within seconds, with light output that is proportional to luciferase activity or ATP concentrations. DMNPE-caged luciferin readily crosses cell membranes, allowing more efficient delivery of luciferin into intact cells. Once the caged luciferin is inside the cell, active luciferin can be released either instantaneously by a flash of UV light, or continuously by the action of endogenous intracellular esterases found in many cell types.

Photoactivatable fluorescent dyes, which are generally colorless and nonfluorescent until photolyzed with UV light, are particularly useful for investigating cell lineage and for spatiotemporal interrogation of fluid flows. In addition to CMNB-caged fluorescein (F7103), we prepare the succinimidyl ester of CMNB-caged carboxyfluorescein (C20050), which can be used to attach the caged fluorophore to primary amine groups of a variety of biomolecules. CMNB-caged carboxyfluorescein succinimidyl ester is a key starting material in the preparation of probes for super-resolution photoactivation microscopy. Furthermore, caged fluorescein probes are immunochemically cryptic; i.e., the probe is immunoreactive with anti–fluorescein/Oregon Green dye antibodies (Anti-Dye and Anti-Hapten Antibodies—Section 7.4) after but not before photoactivation (Figure 5.3.9).

Figure 5.3.9 Schematic representation of photoactivated fluorescence combined with sample masking. Initially, no fluorescence is observed from samples stained with a CMNB-caged fluorescein-labeled secondary detection reagent (A). The desired mask is then placed over the sample (B), after which the sample is exposed to UV light. The mask is then removed; fluorescein molecules present in the unmasked portion of the sample are uncaged by the UV light and fluoresce brightly when viewed with the appropriate filters (C). Uncaged fluorescein may now also serve as a hapten for further signal amplification using our anti–fluorescein/Oregon Green dye antibody (Anti-Dye and Anti-Hapten Antibodies—Section 7.4). For example, probing with the anti–fluorescein/Oregon Green dye antibody followed by staining with the Alexa Fluor 594 goat anti–mouse IgG antibody can be used to change the color of the uncaged probe to red fluorescent (D).

Using organic synthesis methods, researchers can cage a diverse array of molecules. One of the preferred caging groups is the 1-(4,5-dimethoxy-2-nitrophenyl)ethyl (DMNPE) ester. Because the diazoethane precursor to DMNPE esters is unstable, we recommend starting with the hydrazone precursor (4,5-dimethoxy-2-nitroacetophenone hydrazone) and using manganese (IV) oxide (MnO₂ ) for oxidation to generate the reactive diazoethane. A wide range of compounds containing a weak oxy acid (with a pK_a between 3 and 7), including carboxylic acids, phenols and phosphates, should react with the diazoethane to form the DMNPE-caged analogs (Figure 5.3.10). If the target molecule has multiple reactive centers, this synthetic method produces a mixture of products that require chromatographic separation.

Figure 5.3.10 Caging of a carboxylic acid using the hydrazone precursor of DMNPE, 4,5-dimethoxy-2-nitroacetophenone hydrazone.