武汉大学生命科学学院：《分子生物学》（英文版）Nucleic Acid Hybridization

Linker length and sequence have been shown to have a major impact as well. Very short linkers(4-10 atoms) inhibit incorpora tion and affect hybridization(Haralambidis, Chai, and Tregear, 1987). Above a certain linker length(>20 atoms), incorporation and hybridization are impaired, and a model of steric hindrance has been postulated to explain this effect(Zhu, Chao, and Waggoner, 1994) With few exceptions, radioactive and nonradioactive probes should not be internally labeled to 100% completion Strong beta emitters (P, P) will degrade extensively labeled probes. A random primer-generated probe labeled withP-dCTP to a spe cific activity of 10 cpm/ug will have an average size of about 300 to 500 nucleotides immediately after labeling. After storage at 4C for 24 hours, the average size falls to 100 nucleotides or less (Amersham Pharmacia Biotech, unpublished observations ). Max imally diluting the probe immediately after labeling can reduce this radiolytic degradation. High densities of large nonradioactive tags can interfere with duplex formation and strand extension due to steric hindrance (Zhu and Waggoner, 1997; Lee et al., 1992; Day et aL., 1990; Mineno et al., 1993). Although linker chains that connect label to nucleotide are designed to minimize interference(Petrie et al 1991), steric hindrance cannot be completely circumvented. High label densities can also cause quenching. Quenching effects for fluorescein densities greater than l in 10 have been described(Makrigiorgos, Chakrabarti, and Mahmood, 1998) Manufacturers of nonradioactive labeling kits optimize protocols to avoid interference from the label. Consult your manufacturer before you alter a recommended procedure. RADIOACTIVE AND NONRADIOACTIVE LABELING STRATEGIES COMPARED The decision to apply radioactive or nonradioactive labeling and detection systems can be based on issues of sensitivity, high throughput, cost, safety, and ease of use, to name a few criteria discussed in this chapter. While it is feasible on paper to evaluate your research needs against these criteria, the decision must usually be determined at the lab bench. In lieu of the need to test individual systems, several studies compared the sensitivity of nonradioactive probes to the P gold standard: Yang et al. (1999) Nucleic Acid Hybridization 409

Linker length and sequence have been shown to have a major impact as well. Very short linkers (4–10 atoms) inhibit incorpora￾tion and affect hybridization (Haralambidis, Chai, and Tregear, 1987). Above a certain linker length (>20 atoms), incorporation and hybridization are impaired, and a model of steric hindrance has been postulated to explain this effect (Zhu, Chao, and Waggoner, 1994). Overlabeling With few exceptions, radioactive and nonradioactive probes should not be internally labeled to 100% completion. Strong beta￾emitters (32P, 33P) will degrade extensively labeled probes. A random primer-generated probe labeled with 32P-dCTP to a spe￾cific activity of 109 cpm/mg will have an average size of about 300 to 500 nucleotides immediately after labeling.After storage at 4°C for 24 hours, the average size falls to 100 nucleotides or less (Amersham Pharmacia Biotech, unpublished observations). Max￾imally diluting the probe immediately after labeling can reduce this radiolytic degradation. High densities of large nonradioactive tags can interfere with duplex formation and strand extension due to steric hindrance (Zhu and Waggoner, 1997; Lee et al., 1992; Day et al., 1990; Mineno et al., 1993). Although linker chains that connect label to nucleotide are designed to minimize interference (Petrie et al., 1991), steric hindrance cannot be completely circumvented. High label densities can also cause quenching. Quenching effects for fluorescein densities greater than 1 in 10 have been described (Makrigiorgos, Chakrabarti, and Mahmood, 1998). Manufacturers of nonradioactive labeling kits optimize protocols to avoid interference from the label. Consult your manufacturer before you alter a recommended procedure. RADIOACTIVE AND NONRADIOACTIVE LABELING STRATEGIES COMPARED The decision to apply radioactive or nonradioactive labeling and detection systems can be based on issues of sensitivity, high￾throughput, cost, safety, and ease of use, to name a few criteria discussed in this chapter. While it is feasible on paper to evaluate your research needs against these criteria, the decision must usually be determined at the lab bench. In lieu of the need to test individual systems, several studies compared the sensitivity of nonradioactive probes to the 32P gold standard: Yang et al. (1999), Nucleic Acid Hybridization 409

Plath, Peters, and Einspanier (1996), Nass and Dickson (1995), Moore and Margolin (1993), Puchhammer-Stoeckl, Heinz, and Kunz(1993), Engler-Blum et al. (1993), Bright et al.(1992), Kanematsu et al. (1991), Lion and Haas(1990), Jiang, Estes, and metcalf (1987), Te al.(1998), Holtke et al. (1992) Pollard-Knight et al. (1990), Hill and Crampton(1994), Dubitsky, Brown, and Brandwein(1992), and Nakagami et al. (1991) What Are the Criteria for Considering direct over Indirect Nonradioactive Labeling Strategies? Direct labeling strategies utilize probes that are directly conju gated to a dye or an enzyme, which generates the detection signal Indirect labeling systems utilize probes that contain a hapten that will bind to a secondary agent generating the detection signal the probe itself does not generate signal. Typical secondary agents are dye-or enzyme-linked antibodies, and enzyme-linked avidin exes c Comparing the sensitivities of indirect and direct strategies is a ifficult process. The fluorescent tags or dyes incorporated directly into probes usually have lower sensitivity. Detection limits wi vary with tag or label incorporation efficiency, amplification level introduced by the secondary agent, and amplification level added by substrate or dye. In one instance, simply increasing the dura tion of the labeling reaction within a direct labeling strategy pro- duced more sensitive probes than an indirect labeling strategy which in previous experiments had produced the more sensitive probe(Herzer, P, unpublished observations) Comparisons are further complicated because direct indirect labeling or detection strategies demand very different hybridization, washing, and detection procedures. Additionally the performances of these different strategies can vary with the size and structure of a probe or template from which the probe will be synthesized, further complicating any prediction of performance Manufacturers of labeling and detection systems can usually provide sensitivity comparisons of their different products that are qualitatively, if not always quantitatively reproducible Flexibility a directly labeled probe can be detected after hybridization and ashes; no further blocking or antibody steps are required. Direct nonradioactive techniques limit choices for hybridization, wash 410 Herzer and Englert

Plath, Peters, and Einspanier (1996), Nass and Dickson (1995), Moore and Margolin (1993), Puchhammer-Stoeckl, Heinz, and Kunz (1993), Engler-Blum et al. (1993), Bright et al. (1992), Kanematsu et al. (1991), Lion and Haas (1990), Jiang, Estes, and Metcalf (1987), Tenberge et al. (1998), Holtke et al. (1992), Pollard-Knight et al. (1990), Hill and Crampton (1994), Dubitsky, Brown, and Brandwein (1992), and Nakagami et al. (1991). What Are the Criteria for Considering Direct over Indirect Nonradioactive Labeling Strategies? Direct labeling strategies utilize probes that are directly conju￾gated to a dye or an enzyme, which generates the detection signal. Indirect labeling systems utilize probes that contain a hapten that will bind to a secondary agent generating the detection signal; the probe itself does not generate signal. Typical secondary agents are dye- or enzyme-linked antibodies, and enzyme-linked avidin complexes. Sensitivity Comparing the sensitivities of indirect and direct strategies is a difficult process.The fluorescent tags or dyes incorporated directly into probes usually have lower sensitivity. Detection limits will vary with tag or label incorporation efficiency, amplification level introduced by the secondary agent, and amplification level added by substrate or dye. In one instance, simply increasing the dura￾tion of the labeling reaction within a direct labeling strategy pro￾duced more sensitive probes than an indirect labeling strategy which in previous experiments had produced the more sensitive probe (Herzer, P., unpublished observations). Comparisons are further complicated because direct and indirect labeling or detection strategies demand very different hybridization, washing, and detection procedures.Additionally the performances of these different strategies can vary with the size and structure of a probe or template from which the probe will be synthesized, further complicating any prediction of performance. Manufacturers of labeling and detection systems can usually provide sensitivity comparisons of their different products that are qualitatively, if not always quantitatively reproducible. Flexibility A directly labeled probe can be detected after hybridization and washes; no further blocking or antibody steps are required. Direct, nonradioactive techniques limit choices for hybridization, wash 410 Herzer and Englert

buffer, and detection options. Optimization of your signal-to-noi ratio might be more difficult. Indirect nonradioactive detection systems are usually compatible with the common hybridization and wash buffers, but subsequent antibody incubation and detec- tion steps can be difficult to optimize What Is the Storage Stability of Labeled Probes? Radioactive The effect of high and low emissions from radioactive labels, and methods to minimize their impact are discussed in Chapter 6 Ideally probes labeled with P, P, orS should be prepared fresh for each experiment. If you choose to store a radiolabeled probe, the unincorporated label should be removed prior to storage Damage from radioactive emissions can be minimized by dilution and the addition of free radical scavengers such hanol or reducing agents, but the probe must then be re-purified before reuse. As discussed in Chapter 6, it is crucial to fast-freeze radio- labeled probes to avoid complications from clustering Nonradioactive The chemical nature of the tag will dictate specific storage con ditions, but in general, nonradioactively labeled probes may be aliquoted and stored in the dark in-20 or-70C non-frost-fre freezers. Multiple freeze-thaws should be avoided Stability varie depending on storage buffer formulation and the nature Nucleic acids labeled by direct crosslinking of enzymes are sup- posed to be stable if stored in 50% glycerol at-20oC for several months, but this cannot be guaranteed. Since it is difficult to quan titate the remaining enzyme activity after storage, it is recom mended that fresh probes be prepared to ensure that results over time are comparable. For probes labeled with chromophores or fluorophores, it is crucial to contact the manufacturer for the most contemporary storage information In a system dependent upon a enzyme-labeled antibody, the storage conditions must maintain the integrity of the antibody, the enzyme and the probe If you plan to apply any probe (radioactive or not)over the long term, a positive control that can be used to evaluate the probe effectivenss is highly recommended. Probe stability is also a func- tion of the required sensitivity. If an old preparation of a labeled probe generates the desired signal, the probe was sufficiently Nucleic Acid Hybridization 41 I

buffer, and detection options. Optimization of your signal-to-noise ratio might be more difficult. Indirect nonradioactive detection systems are usually compatible with the common hybridization and wash buffers, but subsequent antibody incubation and detec￾tion steps can be difficult to optimize. What Is the Storage Stability of Labeled Probes? Radioactive The effect of high and low emissions from radioactive labels, and methods to minimize their impact are discussed in Chapter 6. Ideally probes labeled with 32P, 33P, or 35S should be prepared fresh for each experiment. If you choose to store a radiolabeled probe, the unincorporated label should be removed prior to storage. Damage from radioactive emissions can be minimized by dilution and the addition of free radical scavengers such as ethanol or reducing agents, but the probe must then be re-purified before reuse. As discussed in Chapter 6, it is crucial to fast-freeze radio￾labeled probes to avoid complications from clustering. Nonradioactive The chemical nature of the tag will dictate specific storage con￾ditions, but in general, nonradioactively labeled probes may be aliquoted and stored in the dark in -20 or -70°C non-frost-free freezers. Multiple freeze–thaws should be avoided. Stability varies depending on storage buffer formulation and the nature of the tag (e.g., fluorescent label and enzyme), and can vary from one month to one year. Nucleic acids labeled by direct crosslinking of enzymes are sup￾posed to be stable if stored in 50% glycerol at -20°C for several months, but this cannot be guaranteed. Since it is difficult to quan￾titate the remaining enzyme activity after storage, it is recom￾mended that fresh probes be prepared to ensure that results over time are comparable. For probes labeled with chromophores or fluorophores, it is crucial to contact the manufacturer for the most contemporary storage information. In a system dependent upon a enzyme-labeled antibody, the storage conditions must maintain the integrity of the antibody, the enzyme and the probe. If you plan to apply any probe (radioactive or not) over the long term, a positive control that can be used to evaluate the probe’s effectivenss is highly recommended. Probe stability is also a func￾tion of the required sensitivity. If an old preparation of a labeled probe generates the desired signal, the probe was sufficiently stable. Nucleic Acid Hybridization 411

Should the Probe Previously Used within the Hybridization Solution of an Earlier Experiment Be Applied in a New Experiment? Does Sufficent Probe remain? lost blotting applications require and use probe in excess over target and depending on the amount of probe bound to the blot, and on decay and decomposition effects, sufficient probe might remain in the hybridization solution. a dot blot of a dilution series of the target DNA can determine if sufficient probe remains. Label potency Consider the many issues regarding the lifetime and storage stability of labels and tags mentioned above. I(S.H. have suc- cessfully reused random primer-labeled probes up to five day after the initial hybridization experiment. Storage of radiolabeled probes at -20oC in hybridization buffer for a few days or at 4C overnight is usually not problematic. Reuse of radiolabeled probe is not recommended for high-sensitivity and/or quantitative appli cations. The storage issues discussed above for radioactive and nonradioactive labels should also be considered here It may be worthwhile to re-purify some probes from the buffer for optimal storage Peptide nucleic acids(PNA) probes are an example where the probe expense may justify the expense of re- purification How Should a probe be denatured for reuse? Probes are stored in hybridization buffer prior to reuse. Such buffers may contain components that will aid the denaturation of probe(e. g, formamide)so that boiling is not required. Heating to temperatures above Tm is sufficient, since at Tm half of the nucleic acid is denatured. Boiling can also destroy buffer com- ponents such as blocking reagents, SDS, volume excluders, and label. Heating the hybridization buffer containing the probe to temperatures of 10 to 20C above the hybridization temperature would be ideal, but 20 to 30oC below the boiling point has to suffice Is It Essential to Determine the Incorporation Efficiency of Every Labeling Reaction? Labeling reactions are straightforward, and with the advent of commercial labeling kits, unlikely to fail. Hence we do not measure Inco rporation efficiency of every probe that we label We only begin to question labeling efficiency when hybridization 412 Herzer and Englert

Should the Probe Previously Used within the Hybridization Solution of an Earlier Experiment Be Applied in a New Experiment? Does Sufficent Probe Remain? Most blotting applications require and use probe in excess over target and depending on the amount of probe bound to the blot, and on decay and decomposition effects, sufficient probe might remain in the hybridization solution.A dot blot of a dilution series of the target DNA can determine if sufficient probe remains. Label Potency Consider the many issues regarding the lifetime and storage stability of labels and tags mentioned above. I (S.H.) have suc￾cessfully reused 32P random primer-labeled probes up to five days after the initial hybridization experiment. Storage of radiolabeled probes at -20°C in hybridization buffer for a few days or at 4°C overnight is usually not problematic. Reuse of radiolabeled probes is not recommended for high-sensitivity and/or quantitative appli￾cations. The storage issues discussed above for radioactive and nonradioactive labels should also be considered here. It may be worthwhile to re-purify some probes from the buffer for optimal storage. Peptide nucleic acids (PNA) probes are an example where the probe expense may justify the expense of re-purification. How Should a Probe be Denatured for Reuse? Probes are stored in hybridization buffer prior to reuse. Such buffers may contain components that will aid the denaturation of probe (e.g., formamide) so that boiling is not required. Heating to temperatures above Tm is sufficient, since at Tm half of the nucleic acid is denatured. Boiling can also destroy buffer com￾ponents such as blocking reagents, SDS, volume excluders, and label. Heating the hybridization buffer containing the probe to temperatures of 10 to 20°C above the hybridization temperature would be ideal, but 20 to 30°C below the boiling point has to suffice. Is It Essential to Determine the Incorporation Efficiency of Every Labeling Reaction? Labeling reactions are straightforward, and with the advent of commercial labeling kits, unlikely to fail. Hence we do not measure incorporation efficiency of every probe that we label. We only begin to question labeling efficiency when hybridization 412 Herzer and Englert

results are unsatisfactory, a point at which it might be too late to determine incorporation efficiency Before skipping any control steps, consider the implications Minimally, measure incorporation efficiency when working with a new technique, a new probe, a new protocol, or a new kit. Radio- labeled probes need to be purified or at least Trichloroacetic acid (TCA) precipitated to determine labeling efficiency, as discussed in Chapter 7, "DNA Purification. "Determining the efficiency of nonradioactive labeling reactions can be more time-consuming, often involving dot blots and/or scanning of probe spots. Follow manufacturer recommendations to determine labeling efficiency of nonradioactive probes. Is It Necessary to Purify Every Probe Unincorporated nucleotides, enzyme, crosslinking reagents, buffer components, and the like, may cause high backgrounds or interfere with downstream experiments. Hybridization experi- ments where the volume of the probe labeling reaction is negli gible in comparison to the hybridization buffer volume do not always require probe cleanup. If you prefer to minimize these risks, purify the probe away from the reaction components. While there are some labeling procedures (i.e, probes gener- ated by random primer labeling withP-dCTP), where unpurified probe can produce little or no background(Amersham Pharma cia Biotech, unpublished observations), such ideal results can,t be guaranteed for every probe. When background is problematic researchers have the option to repurify the probe preparation Admittedly, this approach wouldn't be of much use if the experi ment producing the background problem required a five day exposure (Purification options are discussed in Chapter 7, " DNA Purification.”) HYBRIDIZATION MEMBRANES AND SUPPORTS What Are the Criteria for Selecting a Support for Your Hybridization Experiment? Beyond the information listed below and your personal experi ence, the most reliable approach to determine if a membrane can be used in your application is to ask the manufacturer for appli cation and or quality control data. Whether a new membrane for mulation will provide you with superior results is a matter that can usually be decided only at the bench, and the results can vary for different sets of targets, probes, and detection strategies. Nucleic Acid Hybridization 4l3

results are unsatisfactory, a point at which it might be too late to determine incorporation efficiency. Before skipping any control steps, consider the implications. Minimally, measure incorporation efficiency when working with a new technique, a new probe, a new protocol, or a new kit. Radio￾labeled probes need to be purified or at least Trichloroacetic acid (TCA) precipitated to determine labeling efficiency, as discussed in Chapter 7, “DNA Purification.” Determining the efficiency of nonradioactive labeling reactions can be more time-consuming, often involving dot blots and/or scanning of probe spots. Follow manufacturer recommendations to determine labeling efficiency of nonradioactive probes. Is It Necessary to Purify Every Probe? Unincorporated nucleotides, enzyme, crosslinking reagents, buffer components, and the like, may cause high backgrounds or interfere with downstream experiments. Hybridization experi￾ments where the volume of the probe labeling reaction is negli￾gible in comparison to the hybridization buffer volume do not always require probe cleanup. If you prefer to minimize these risks, purify the probe away from the reaction components. While there are some labeling procedures (i.e., probes gener￾ated by random primer labeling with 32P-dCTP), where unpurified probe can produce little or no background (Amersham Pharma￾cia Biotech, unpublished observations), such ideal results can’t be guaranteed for every probe. When background is problematic, researchers have the option to repurify the probe preparation. Admittedly, this approach wouldn’t be of much use if the experi￾ment producing the background problem required a five day exposure. (Purification options are discussed in Chapter 7, “DNA Purification.”) HYBRIDIZATION MEMBRANES AND SUPPORTS What Are the Criteria for Selecting a Support for Your Hybridization Experiment? Beyond the information listed below and your personal experi￾ence, the most reliable approach to determine if a membrane can be used in your application is to ask the manufacturer for appli￾cation and or quality control data. Whether a new membrane for￾mulation will provide you with superior results is a matter that can usually be decided only at the bench, and the results can vary for different sets of targets, probes, and detection strategies. Nucleic Acid Hybridization 413