- Email: [email protected]
Rapid isolation of total RNA from mammalian tissues
ANALYTICAL
BIOCHEMISTRY
174,658-66 I(1 988)
Rapid Isolation of Total RNA from Mammalian MARK EMMETT
AND BARBARA
Research Department, Pharmaceutical Research, CIBA-GEIGY
Tissues
PETRACK’ Corporation, Summit. New Jersey 07901
Received January 27, 1988 A rapid procedure for the isolation of total RNA from small amounts of mammalian tissue (35 to 150 mg) is described. Tissues were homogenized in the presence of RNase inhibitors but in the absence of strong detergents. Contaminants were removed by phenol/chloroform extraction and Sephadex column chromatography. Total RNAs were precipitated with ethanol and sodium acetate. The RNAs isolated were intact and suitable for mRNA quantitation via Northern blot or slot-blot analyses. This procedure isolates total RNAs in high yield and purity, without CsCl ultracentrifugation, and is especially useful when mRNAs must be quantitated from many samples. 0 1988 Academic Press, Inc. KEY WORDS: total RNA; RNA isolation; Sephadex columns; RNA purification.
By coupling mRNA quantitation with other biochemical techniques, it is possible to determine if a peptide is regulated by synthesis or by degradation. A crucial step in mRNA quantitation is the isolation of intact total RNA from tissue samples. The guanidinium thiocyanate/CsCl density gradient ultracentrifugation procedure of Chirgwin et al. (1) is currently the most widely used technique for the isolation of RNA, yielding high quality, protein-free RNA. But the procedure requires an ultracentrifuge and utilizes a large amount of tissue, and rotor capacity limits the number of samples that can be processed. A much shorter extraction procedure involves phenol/chloroform extractions in the presence of RNase inhibitors (2). RNA isolated by this procedure is not of the same quality as that obtained by the CsCl gradient method because of protein and phenol contamination (based on OD260/OD280 ratios). Multiple phenol/chloroform extractions followed by chloroform extractions increase the purity but decrease the yield of RNA.
A modification of the phenol extraction procedure, coupled with Sephadex G-50 column chromatography (phenol/Sephadex G50 extraction), has been developed, yielding total RNA from small quantities of tissue (35-150 mg) with OD&ODZK, ratios of approximately 1.8. This technique allows more samples to be processed than with CsCl ultracentrifugation, and with similar yields. Recently, another noncentrifugation procedure has been reported for the isolation of total RNA in high yield and purity (3). MATERIALS
Preparations of reagents. All reagents were molecular biology grade or of the highest purity available. Sterile plastic labware was used whenever possible. Polypropylene tubes were autoclaved for 30 min. Gloves were worn at all times. All buffers were made in diethylpyrocarbonate (DEPC)’ treated 3X glassdistilled water (0.1% DEPC, boiled for 30 min and then autoclaved for 30 min). Homogenization buffer, made fresh before use, consisted of 1.4 ml of 3.0 M sodium acetate, pH
’ To whom all correspondence should be addressed, at CIBA-GEIGY Corp., Room J-108, 556 Morris Ave., Summit, NJ 07901. 0003-2697/88 $3.00 Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.
AND METHODS
* Abbreviation used: DEPC, diethylpyrocarbonate. 658
ISOLATION
OF TOTAL
659
RNA
TABLE 1 YIELD
AND PURITY
OF TOTAL
RNA
ISOLATED
FROM
BRAIN REGIONS
AND LIVER
Guanidinium/CsCl ultracentrifugation
Phenol/Sephadex extraction Tissue Striatum Hypothalamus
Yield bug/g)
Purity (OJho/OD28o)
912.00 f 67.24 (37) 112.50 f 36.19
1.76 kO.012 (56) 1.8 1 + 0.009 (56) 1.83 + 0.009
(36) Cortex Hippocampus Liver
781.25 f 38.2 (34) 929.50 + 44.85 (34) 4729.50 + 29 1.1
(8)
(56) 1.79 * 0.01 (56) 1.91 ~0.009
(8)
Yield Wg) 588
Purity (OD2em/OD28o) 1.90
(2)
(2)
690
1.98
(2)
(2)
414
1.83
(2)
(2)
1067
1.84
(2)
(2)
Note. Total RNA was isolated from regions of rat brain and liver as described under Materials and Methods for the phenol/Sephadex extraction procedure and as reported by Chirgwin et al. (1) for the guanidinium/CsCl ultracentrifugation method. The yield was calculated from the OD lco, taking an OD of 1.0 equal to 40 &ml. The values shown are the mean f SE for the number of determinations shown in parentheses for the new procedure and the average of two separate determinations for isolation by guanidinium/CsCl ultracentrifugation.
6.0,0.28mlof5.0MEDTA,pH8.5-9.0,0.28 ml of 200 nM vanadyl ribonucleoside complexes, 0.35 ml of heparin solution (5000 units/ml), and 9.15 ml of DEPC water. Buffered phenol was prepared by melting redistilled phenol at 60°C adding 8-hydroxyquinoline to 0.5%, extracting three times (in dim light) at 4°C with an equal volume of 1 .O M Tris-HCl, pH 8.0, and then with 0.1 M Tris-HCl, pH 8.0, until the pH of the aqueous phase was >7.6. The phenol was stored under 0.1 M Tris-HCl buffer, pH 8.0, at 4°C in a brown bottle and used within 10 days. RNase-free, prepacked, preswollen Sephadex G-50 columns (Boeringer-Mannheim, Catalog No. 100-4 11) were utilized for column chromatography. RNA isolation. Rats were killed by decapitation. The brains were removed and dissected as rapidly as possible. Brain regions were placed in sterilized preweighed 12 X 75mm polypropylene tubes, frozen in liquid nitrogen, and stored at -70°C until use. Just before homogenization, the tissues were removed from the -70°C freezer, placed on dry
ice, and rapidly weighed. Each frozen tissue (35 to 150 mg) was homogenized on ice in 0.85 ml of homogenization buffer for 10 s using a Polytron set at the No. 7 speed, followed immediately by addition of 100 bl of 10% SDS. The tubes were vortexed and put on ice. (The number of samples was limited to 12 or less per isolation; three to four sets could be processed per day). Buffered phenol (1 ml) was added to each tube and vortexed until homogeneous. One milliliter of chloroform/ isoamyl alcohol (24:l) was added to each tube and vortexed until homogeneous. The tubes were incubated in a 55°C water bath for 5 min (the chemical reaction between the vanadyl ribonucleoside complexes and the 8-hydroxyquinoline turns the solution from grey to black), removed, vortexed, and placed on ice for 5 min. The tubes were then centrifuged in a Sorvall SM-24 rotor at 17,000 rpm (35,500g) at 4°C for 30 min, removed from the rotor, and placed on ice. The upper aqueous layer ( 1.O- 1.2 ml) was removed and placed directly onto a predrained RNase-free Sephadex G-50 column.
660
EMMETT
AND PETRACK
total RNA was precipitated overnight at -70°C. After precipitation, the RNA was pelleted by centrifugation in a Beckman microfuge 12 at top speed, at 4°C for 30 min. The supernatants were decanted and the RNA pellets were dried under vacuum in a cooled Savant Speed Vat for 40 min. The RNA pellets from each tissue were resuspended in 50 ~1 of sterile DEPC-treated 3X glass-distilled water. Total RNA was quantitated by uv spectroscopy. Samples can be stored in the DEPC water at -70°C for extended periods of time.
28s
A.
1234
18s
B.
FIG. 1. Electrophoresis of tota RNA (20 pg). isolated from rat cerebral cortex by the phenol/Sephedex G-50 extraction procedure, before (lane 2) and after (lane 4) digestion with RNase. RNA ladder before (lane 1) and after (lane 3) RNase-treatment are also shown. The samples were eiectraphoresed in formaldehyde (6%)-agarose (0.85%) gel for 6 h at 28 V. The gels were washed and stained with ethidium bromide.
The eluent was collected in a sterile 1.9 ml (2.2 ml is better) polypropylene microfuge tube which was supplemented with 100 ~1 of 3 M sodium acetate. If the flow stopped, the liquid was pushed through with air using a rubbei Pasteur pipet bulb. The column was washed with 0.2 ml of solution containing 10 mM Tris, pH 8.0, 1 mM EDTA, and 0.1 M NaCI. The mixed eluent was divided into two microfuge tubes; each tube was filled with 100% ethanol (approx 1.2 ml, -2O”C), and
?50-
FIG. 2. Northern blot autoradiogram of peptide mRNA from rat brain regions by the phenol/Sephadex G-SO extraction procedure. Total RNA samples (17 pg) were denatured, electrophoresed on agarose-formaldehyde, transferred to nitrocellulose by capillary action, and affixed by baking for 2 hrs, at 80°C under vacuum. The filters were hybridized with a probe for preproenkephalin, washed, and rehybridized with a probe for preprocholecystokinin; both probes were labeled with [“PIdCTP. Each filter was exposed to Kodak XOMAT-RP film with a Cronex intensifying screen for 14 h. The autoradiogram identified a mRNA of approximately 1.4 kb when hybridized with the preproenkephalin probe (A) and of approximately 750 bases with the preprocholecystokinin probe (B). These values are consistent with published reports (6,7). The brain regions analyzed were 1, hypothalamus; 2, hippocampus; 3, frontal cortex; and 4, brain stem.
ISOLATION
OF TOTAL
RESULTS AND DISCUSSION
Table 1 demonstrates the high yield and purity (based on OD&OD280 ratios) of total RNA isolated by the described procedure from various brain regions and liver samples. Both the yield and purity are similar to those obtained by the guanidinium thiocyanate/ CsCl ultracentrifugation procedure. Both isolation procedures yield similar patterns on formaldehyde-agarose (0.8%) gel electrophoresis, visualized with ethidium bromide. A sample of total RNA isolated from rat cerebral cortex by the rapid procedure was incubated with RNase. Figure 1 illustrates the ethidium bromide-stained gel, with the typical 28s and 18s ribosomal RNAs; both bands were absent in the samples that had been incubated with RNase. Total RNA isolated by the phenol/Sephadex G-50 extraction procedure has been used in our laboratory to quantitate three different neuropeptide mRNAs from specific regions of rat brain via Northern blots and slot blot analyses. Figure 2 shows a Northern blot of total RNA from various brain regions hybridized with two different 32P-labeled rat cDNA probes, one encoding preproenkephalin, the other encoding preprocholecystokinin. Each probe identified a mRNA of the correct size and there was little or no smearing on the autoradiograms (due to RNA degradation). With this procedure, enough total RNA was isolated from each of the dissected brain regions (striatum, hypothalamus, frontal cortex, and hippocampus) of one rat to measure the mRNA level in each tissue via Northern blotting and slot blotting. Liver RNA (used as a negative control) did not hybridize with either probe. RNA degradation was minimal due to the presence of three RNase inhibitors: vanadyl ribonucleoside complexes, heparin (4), and 8-hydroxyquinoline (5). Homogenization performed in the absence of detergents avoids nuclear rupture, minimizing DNA contami-
661
RNA
nation. Phenol extraction with a high concentration of 8-hydroxyquinoline denatures proteins and removes most of the vanadyl ribonucleoside complexes which interfere with uv absorbance. The high speed centrifugation resulted in a sharp interphase. Occasionally, the organic/aqueous interphase was fairly large (interphase size varied with tissue type and amount), but usually was sharpened by additional centrifugation. The Sephadex column collected most of the denatured protein on top, retarded the phenol, oligonucleotides, and excess vanadyl ribonucleoside complexes, and excluded the total RNAs in the void volume. Chloroform was removed during drying under vacuum. The isolation procedure (up to ethanol precipitation) may be completed in approximately 2 h. Compared with CsCl ultracentrifugation, the phenol/Sephadex G-50 procedure enables isolation of total RNA from many more samples in less time and with similar yield and purity (based on protein contamination). This procedure is especially useful when mRNAs must be quantitated from many different samples by Northern blot analysis. REFERENCES 1. Chirgwin, J. M., Ptzybyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979) Biochemistry 18, 5296 5299. 2. Lizardi, P. M. (1983) in Methods in Enzymology (Fleisher, S., and Fleisher, B., Eds.), Vol. 96, pp. 24-38, Academic Press, New York. 3. Chomczynski, P., and Sacchi, N. (1987) Anal. Biothem. 162,156-159. 4. Berger, S. L., and Birkenmeier, C. S. (1979) Biochemistry 18,5 143-5 149. 5. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, p. 438, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 6. Tang, F.. Costa, E., and Schwartz, J. P. (1983) Proc. Natl. Acad. Sci. USA 80,3841-3844. 7. Yoshikawa, K., Williams, C., and Sabol, S. L. ( 1984) J. Biol. Chem. 259,14301-14308.
BIOCHEMISTRY
174,658-66 I(1 988)
Rapid Isolation of Total RNA from Mammalian MARK EMMETT
AND BARBARA
Research Department, Pharmaceutical Research, CIBA-GEIGY
Tissues
PETRACK’ Corporation, Summit. New Jersey 07901
Received January 27, 1988 A rapid procedure for the isolation of total RNA from small amounts of mammalian tissue (35 to 150 mg) is described. Tissues were homogenized in the presence of RNase inhibitors but in the absence of strong detergents. Contaminants were removed by phenol/chloroform extraction and Sephadex column chromatography. Total RNAs were precipitated with ethanol and sodium acetate. The RNAs isolated were intact and suitable for mRNA quantitation via Northern blot or slot-blot analyses. This procedure isolates total RNAs in high yield and purity, without CsCl ultracentrifugation, and is especially useful when mRNAs must be quantitated from many samples. 0 1988 Academic Press, Inc. KEY WORDS: total RNA; RNA isolation; Sephadex columns; RNA purification.
By coupling mRNA quantitation with other biochemical techniques, it is possible to determine if a peptide is regulated by synthesis or by degradation. A crucial step in mRNA quantitation is the isolation of intact total RNA from tissue samples. The guanidinium thiocyanate/CsCl density gradient ultracentrifugation procedure of Chirgwin et al. (1) is currently the most widely used technique for the isolation of RNA, yielding high quality, protein-free RNA. But the procedure requires an ultracentrifuge and utilizes a large amount of tissue, and rotor capacity limits the number of samples that can be processed. A much shorter extraction procedure involves phenol/chloroform extractions in the presence of RNase inhibitors (2). RNA isolated by this procedure is not of the same quality as that obtained by the CsCl gradient method because of protein and phenol contamination (based on OD260/OD280 ratios). Multiple phenol/chloroform extractions followed by chloroform extractions increase the purity but decrease the yield of RNA.
A modification of the phenol extraction procedure, coupled with Sephadex G-50 column chromatography (phenol/Sephadex G50 extraction), has been developed, yielding total RNA from small quantities of tissue (35-150 mg) with OD&ODZK, ratios of approximately 1.8. This technique allows more samples to be processed than with CsCl ultracentrifugation, and with similar yields. Recently, another noncentrifugation procedure has been reported for the isolation of total RNA in high yield and purity (3). MATERIALS
Preparations of reagents. All reagents were molecular biology grade or of the highest purity available. Sterile plastic labware was used whenever possible. Polypropylene tubes were autoclaved for 30 min. Gloves were worn at all times. All buffers were made in diethylpyrocarbonate (DEPC)’ treated 3X glassdistilled water (0.1% DEPC, boiled for 30 min and then autoclaved for 30 min). Homogenization buffer, made fresh before use, consisted of 1.4 ml of 3.0 M sodium acetate, pH
’ To whom all correspondence should be addressed, at CIBA-GEIGY Corp., Room J-108, 556 Morris Ave., Summit, NJ 07901. 0003-2697/88 $3.00 Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.
AND METHODS
* Abbreviation used: DEPC, diethylpyrocarbonate. 658
ISOLATION
OF TOTAL
659
RNA
TABLE 1 YIELD
AND PURITY
OF TOTAL
RNA
ISOLATED
FROM
BRAIN REGIONS
AND LIVER
Guanidinium/CsCl ultracentrifugation
Phenol/Sephadex extraction Tissue Striatum Hypothalamus
Yield bug/g)
Purity (OJho/OD28o)
912.00 f 67.24 (37) 112.50 f 36.19
1.76 kO.012 (56) 1.8 1 + 0.009 (56) 1.83 + 0.009
(36) Cortex Hippocampus Liver
781.25 f 38.2 (34) 929.50 + 44.85 (34) 4729.50 + 29 1.1
(8)
(56) 1.79 * 0.01 (56) 1.91 ~0.009
(8)
Yield Wg) 588
Purity (OD2em/OD28o) 1.90
(2)
(2)
690
1.98
(2)
(2)
414
1.83
(2)
(2)
1067
1.84
(2)
(2)
Note. Total RNA was isolated from regions of rat brain and liver as described under Materials and Methods for the phenol/Sephadex extraction procedure and as reported by Chirgwin et al. (1) for the guanidinium/CsCl ultracentrifugation method. The yield was calculated from the OD lco, taking an OD of 1.0 equal to 40 &ml. The values shown are the mean f SE for the number of determinations shown in parentheses for the new procedure and the average of two separate determinations for isolation by guanidinium/CsCl ultracentrifugation.
6.0,0.28mlof5.0MEDTA,pH8.5-9.0,0.28 ml of 200 nM vanadyl ribonucleoside complexes, 0.35 ml of heparin solution (5000 units/ml), and 9.15 ml of DEPC water. Buffered phenol was prepared by melting redistilled phenol at 60°C adding 8-hydroxyquinoline to 0.5%, extracting three times (in dim light) at 4°C with an equal volume of 1 .O M Tris-HCl, pH 8.0, and then with 0.1 M Tris-HCl, pH 8.0, until the pH of the aqueous phase was >7.6. The phenol was stored under 0.1 M Tris-HCl buffer, pH 8.0, at 4°C in a brown bottle and used within 10 days. RNase-free, prepacked, preswollen Sephadex G-50 columns (Boeringer-Mannheim, Catalog No. 100-4 11) were utilized for column chromatography. RNA isolation. Rats were killed by decapitation. The brains were removed and dissected as rapidly as possible. Brain regions were placed in sterilized preweighed 12 X 75mm polypropylene tubes, frozen in liquid nitrogen, and stored at -70°C until use. Just before homogenization, the tissues were removed from the -70°C freezer, placed on dry
ice, and rapidly weighed. Each frozen tissue (35 to 150 mg) was homogenized on ice in 0.85 ml of homogenization buffer for 10 s using a Polytron set at the No. 7 speed, followed immediately by addition of 100 bl of 10% SDS. The tubes were vortexed and put on ice. (The number of samples was limited to 12 or less per isolation; three to four sets could be processed per day). Buffered phenol (1 ml) was added to each tube and vortexed until homogeneous. One milliliter of chloroform/ isoamyl alcohol (24:l) was added to each tube and vortexed until homogeneous. The tubes were incubated in a 55°C water bath for 5 min (the chemical reaction between the vanadyl ribonucleoside complexes and the 8-hydroxyquinoline turns the solution from grey to black), removed, vortexed, and placed on ice for 5 min. The tubes were then centrifuged in a Sorvall SM-24 rotor at 17,000 rpm (35,500g) at 4°C for 30 min, removed from the rotor, and placed on ice. The upper aqueous layer ( 1.O- 1.2 ml) was removed and placed directly onto a predrained RNase-free Sephadex G-50 column.
660
EMMETT
AND PETRACK
total RNA was precipitated overnight at -70°C. After precipitation, the RNA was pelleted by centrifugation in a Beckman microfuge 12 at top speed, at 4°C for 30 min. The supernatants were decanted and the RNA pellets were dried under vacuum in a cooled Savant Speed Vat for 40 min. The RNA pellets from each tissue were resuspended in 50 ~1 of sterile DEPC-treated 3X glass-distilled water. Total RNA was quantitated by uv spectroscopy. Samples can be stored in the DEPC water at -70°C for extended periods of time.
28s
A.
1234
18s
B.
FIG. 1. Electrophoresis of tota RNA (20 pg). isolated from rat cerebral cortex by the phenol/Sephedex G-50 extraction procedure, before (lane 2) and after (lane 4) digestion with RNase. RNA ladder before (lane 1) and after (lane 3) RNase-treatment are also shown. The samples were eiectraphoresed in formaldehyde (6%)-agarose (0.85%) gel for 6 h at 28 V. The gels were washed and stained with ethidium bromide.
The eluent was collected in a sterile 1.9 ml (2.2 ml is better) polypropylene microfuge tube which was supplemented with 100 ~1 of 3 M sodium acetate. If the flow stopped, the liquid was pushed through with air using a rubbei Pasteur pipet bulb. The column was washed with 0.2 ml of solution containing 10 mM Tris, pH 8.0, 1 mM EDTA, and 0.1 M NaCI. The mixed eluent was divided into two microfuge tubes; each tube was filled with 100% ethanol (approx 1.2 ml, -2O”C), and
?50-
FIG. 2. Northern blot autoradiogram of peptide mRNA from rat brain regions by the phenol/Sephadex G-SO extraction procedure. Total RNA samples (17 pg) were denatured, electrophoresed on agarose-formaldehyde, transferred to nitrocellulose by capillary action, and affixed by baking for 2 hrs, at 80°C under vacuum. The filters were hybridized with a probe for preproenkephalin, washed, and rehybridized with a probe for preprocholecystokinin; both probes were labeled with [“PIdCTP. Each filter was exposed to Kodak XOMAT-RP film with a Cronex intensifying screen for 14 h. The autoradiogram identified a mRNA of approximately 1.4 kb when hybridized with the preproenkephalin probe (A) and of approximately 750 bases with the preprocholecystokinin probe (B). These values are consistent with published reports (6,7). The brain regions analyzed were 1, hypothalamus; 2, hippocampus; 3, frontal cortex; and 4, brain stem.
ISOLATION
OF TOTAL
RESULTS AND DISCUSSION
Table 1 demonstrates the high yield and purity (based on OD&OD280 ratios) of total RNA isolated by the described procedure from various brain regions and liver samples. Both the yield and purity are similar to those obtained by the guanidinium thiocyanate/ CsCl ultracentrifugation procedure. Both isolation procedures yield similar patterns on formaldehyde-agarose (0.8%) gel electrophoresis, visualized with ethidium bromide. A sample of total RNA isolated from rat cerebral cortex by the rapid procedure was incubated with RNase. Figure 1 illustrates the ethidium bromide-stained gel, with the typical 28s and 18s ribosomal RNAs; both bands were absent in the samples that had been incubated with RNase. Total RNA isolated by the phenol/Sephadex G-50 extraction procedure has been used in our laboratory to quantitate three different neuropeptide mRNAs from specific regions of rat brain via Northern blots and slot blot analyses. Figure 2 shows a Northern blot of total RNA from various brain regions hybridized with two different 32P-labeled rat cDNA probes, one encoding preproenkephalin, the other encoding preprocholecystokinin. Each probe identified a mRNA of the correct size and there was little or no smearing on the autoradiograms (due to RNA degradation). With this procedure, enough total RNA was isolated from each of the dissected brain regions (striatum, hypothalamus, frontal cortex, and hippocampus) of one rat to measure the mRNA level in each tissue via Northern blotting and slot blotting. Liver RNA (used as a negative control) did not hybridize with either probe. RNA degradation was minimal due to the presence of three RNase inhibitors: vanadyl ribonucleoside complexes, heparin (4), and 8-hydroxyquinoline (5). Homogenization performed in the absence of detergents avoids nuclear rupture, minimizing DNA contami-
661
RNA
nation. Phenol extraction with a high concentration of 8-hydroxyquinoline denatures proteins and removes most of the vanadyl ribonucleoside complexes which interfere with uv absorbance. The high speed centrifugation resulted in a sharp interphase. Occasionally, the organic/aqueous interphase was fairly large (interphase size varied with tissue type and amount), but usually was sharpened by additional centrifugation. The Sephadex column collected most of the denatured protein on top, retarded the phenol, oligonucleotides, and excess vanadyl ribonucleoside complexes, and excluded the total RNAs in the void volume. Chloroform was removed during drying under vacuum. The isolation procedure (up to ethanol precipitation) may be completed in approximately 2 h. Compared with CsCl ultracentrifugation, the phenol/Sephadex G-50 procedure enables isolation of total RNA from many more samples in less time and with similar yield and purity (based on protein contamination). This procedure is especially useful when mRNAs must be quantitated from many different samples by Northern blot analysis. REFERENCES 1. Chirgwin, J. M., Ptzybyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979) Biochemistry 18, 5296 5299. 2. Lizardi, P. M. (1983) in Methods in Enzymology (Fleisher, S., and Fleisher, B., Eds.), Vol. 96, pp. 24-38, Academic Press, New York. 3. Chomczynski, P., and Sacchi, N. (1987) Anal. Biothem. 162,156-159. 4. Berger, S. L., and Birkenmeier, C. S. (1979) Biochemistry 18,5 143-5 149. 5. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, p. 438, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 6. Tang, F.. Costa, E., and Schwartz, J. P. (1983) Proc. Natl. Acad. Sci. USA 80,3841-3844. 7. Yoshikawa, K., Williams, C., and Sabol, S. L. ( 1984) J. Biol. Chem. 259,14301-14308.