- Email: [email protected]
GFAP RNA increases during a wasting state in old mice
EXPERIMENTALNEUROLOGY
108,266-268
(1990)
BRIEF COMMUNICATION GFAP RNA increases during a Wasting State in Old Mice JAMES R. GOSS,~ CALEB E. FINCH, Andrus
Gerontology
Center
and Department
of Biological
Sciences,
University
A prolonged wasting condition is often associated with morbidity in older humans. The effects that such a state has on the quality/quantity of RNA are not known. In initial attempts to develop an animal model for premortem wasting, we examined whole brain RNA from mice slowly approaching death from natural causes. Glial fibrillary acidic protein (GFAP) RNA showed a three-fold increase as detected by RNA gel-blot hybridization analysis. Five other RNA sequences were stable under these circumstances. We conclude that brain RNA changes are selective during a degenerating premortem state. Moreover, RNA sequence changes in conditions such as Alzheimer’s disease should be considered in the context of the wasting condition of the individual and may not be due to a direct effect of the disease process. 0 1990 Academic Press, Inc.
In studying chronic neurodegenerative diseases that, like Alzheimer’s disease (AD), arise later in life, a crucial issue is the impact of chronic wasting conditions. AD patients in the terminal phases often are cachectic, anorexic, immobile, and frequently die from oxygen insufficiency secondary to pulmonary infection. Animal models for a prolonged wasting condition are needed to assesssuch influences on brain RNA integrity. Therefore, we examined the brains of mice that were slowly approaching death from natural causesduring aging. We compared total RNA yields, in vitro translation efficiencies, two-dimensional gel profiles, and selected RNA sequences.The only change observed was a threefold increase in the prevalence of RNA coding for GFAP. Female C57BL/GNNia (Charles River; Wilmington, MA) and male C57BL/6J (Jackson Laboratories; Bar Harbor, ME) mice were used. Animals were housed three to five per cage in an aging colony within the Gerontology Center Vivarium. They were maintained on a 12-h/12-h light/dark cycle with food and water available.
1 To whom
0014~4666/90
all correspondence
should
$3.00
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
be addressed. 266
AND
DAVID G. MORGAN
of Southern
California,
Los Angeles,
California
90089-0191
Approximately 70 mice, aged 24 to 35 months, were observed daily for 4 months. Within this time 15% showed signs of a wasting condition and were used for this study. Wasting mice typically failed to ambulate when gently prodded, exhibited slight resting tremors, displayed a hunched posture, and often felt cold to the touch. After 2 consecutive days of these symptoms, mice were killed by cervical dislocation followed by decapitation; whole brain was removed and stored at -70°C. Agematched mice of the same cohort that did not display any of the symptoms were used as controls. A total of seven pairs of animals were examined. Necropsies were performed on all mice. The wasting mice typically revealed gross abnormalities such as tumors of the mesenteric lymph nodes, enlarged spleens, discolored livers, and empty stomachs. Wasting mice typically weighed less than their control mice (26.6 + 2 vs 29.0 + 1.7 g). These abnormalities and weight loss are consistent with previous findings (2, 3). Control mice were healthy in all regards. Whole brain total RNA was extracted using the guanidine thiocyanate and cesium chloride centrifugation method. Eight micrograms of each RNA sample was electrophoresed through a 1% agarose gel containing 6% formaldehyde and transferred to nylon membranes. Membranes were hybridized to one of six cRNA probes. Riboprobes were transcribed from cDNAs inserted into the transcription vector Bluescribe (Stratagene, San Diego, CA) and labeled with [32P]UTP to a specific activity of 10’ cpm/pg. Clones used were as follows: human @amyloid, 1 kb in length corresponding to nt 1794-2871 of the amyloid precursor gene (8); chicken @-tubulin, 1.2 kb in length corresponding to nt 475-1617 (12); mouse GFAP, 1.1 kb in length corresponding to the entire coding region (9); feline glutamic acid decarboxylase (GAD), 1.25 kb in length corresponding to the amino terminal coding region; human somatostatin (SS), 0.45 kb in length corresponding to part of the preprohormone coding region (11); and mouse Thy-l antigen, 0.76 kb in length corresponding to the entire coding region (1). Blots were hybridized overnight at 77°C in 5X SSC containing 0.5% nonfat dry milk, 1% SDS, 10% dextran sulfate, 25 pg/ml poly(A) RNA, 25 pg/ml poly(C) RNA, and
BRIEF
A
Condition
B
c
GFAP -W
267
COMMUNICATION
5-
WHWHWHWH
THY-1
GFAP
GAD
THY-1
B-TUB
88
B-AMY
RNA Sequence8 FIGS. 1A AND 1B. RNA gel-blot hybridizations to glial fibrillary acidic protein and Thy-l antigen. Eight micrograms of mouse whole brain total RNA was loaded into each lane of a 1% agarose, 6% formaldehyde gel. RNA was transferred to a nylon membrane. The immobilized RNA was hybridized with a 32P-labeled cRNA antisense probe to GFAP (A) or Thy-l antigen (B) at 77°C overnight. The blots were washed in successively lower concentrations of SSC and autoradiographs were taken. These blots represent four of the seven pairs of samples. W, wasting mouse; H, healthy age-matched control mouse. FIG. 1C. Effect of wasting state on selected RNA populations. Autoradiographs were analyzed by computer-aided densitometry. The optical densities (OD) of matched pairs of wasting and healthy mice were compared. A ratio of 1 corresponds to no difference between wasting and healthy mice. GFAP, glial fibrillary acidic protein; GAD, glutamic acid decarboxylase; Thy-l, Thy-l antigen; B-Tub, fl-tubulin; SS, Somatostatin; B-Amy, @-amyloid. P < 0.01, Wilcoxon-Mann-Whitney two-sample test.
100 pg/ml sheared salmon sperm DNA. Blots were washed a total of five times with decreasing concentrations of SSC to a final criterion of 0.5X SSC at 77°C. Kodak X-OMAT film was used for autoradiographs. Autoradiographs were analyzed using a computer-aided densitometry program. Four micrograms of total RNA from each sample was translated in vitro using a rabbit reticulocyte lysate system (BRL). A lo-p1 reaction contained 87 mA4 K+, 1 mM Mg+ and used [35S]methionine to label translation products. Radioactive amino acid incorporation into proteins was measured by placing 2 ~1of the reaction mixture into 1 ml of 1 N NaOH/1.5% Hz02 incubating for 10 min at 37°C followed by TCA precipitation. An amount of the translation mixture equal to approximately 100,000 cpm was used for two-dimensional gel analysis (10). Whole brain RNA yields were similar for wasting and control mice (440 + 40 vs 470 -t 80 pg total RNA/g tissue). Ethidium bromide stains of electrophoresed total RNA showed no evidence of RNA degradation. The radioactive amino acid incorporation in the in vitro translation assay was similar for both wasting and control mice (12,600 + 2400 vs 11,700 + 1000 cpm, respectively). Radioactive translation products were resolved on twodimensional gels and showed no qualitative differences in the 200 major spots (data not shown). Northern blot analysis using the six cRNA probes described above showed a large increase in the signal for GFAP RNA (Fig. 1A). There was no change in RNA for GAD, Thy-l (Fig. lB), fl-tubulin, SS, or P-amyloid. Au-
toradiographs were analyzed by computer-aided densitometry and optical density (OD) measurements were made for all samples hybridized to all cRNA probes. A ratio between wasting and control matched pairs was calculated. A three-fold increase was found for GFAP RNA (Fig. 1C). RNA degradation as assessedfrom these Northern blots was minimal. This study shows an increase in the RNA level for GFAP in response to a wasting condition in mice. This increase is above the already elevated GFAP RNA prevalence found in healthy aged mice (5, 6). None of the other five mRNAs were changed. The wasting condition had no effect on the quantity or quality of whole brain total RNA, as measured by total yields, Northern blot band tailing, or in vitro translation efficiencies and products. In AD we found a similar increase in GFAP in frontal and temporal cortex (4). We originally interpreted this as a response to the neurodegeneration associated with the disease but now consider it possible that this increase in GFAP RNA is a response to the premorbid condition associated with AD, rather than a direct effect of the disease process. This suggests that the premortem condition of the tissue donor should be considered when interpreting gene expression differences found in postmortem human brain tissue. In reference to AD, it should be noted that the wasting condition of mice did not affect the amount of P-amyloid precursor protein RNA. However since this study was conducted using whole brain RNA, we cannot rule out small, regionally
268
BRIEF
COMMUNICATION
localized changes in specific messages. Similarly, changes in RNA levels do not necessarily indicate changes in the corresponding protein. To our knowledge, this is the first study describing changes in RNA associated with the premortem state of an animal. We feel that this condition is a variable which should be considered when examining any changes involving gene expression. Furthermore, since many of the wasting animals had visible abnormalities which may influence results, we emphasize the need for necropsies of aged animals in all applicable studies when possible. While it is presently not possible to identify whether any one abnormality visible at necropsy is responsible for the increase in GFAP RNA, we have observed elevated GFAP RNA in one other study which correlated with grossly observable tumors apparent at necropsy (Goss, Finch, and Morgan, unpublished data). Other extraneous variables that may theoretically influence RNA integrity include postmortem interval, medication history, and nutrition. In previous studies, we found no influence of postmortem interval on the quantity or quality of brain RNA up to 48 h in young rats (7). GFAP was not examined in this study. ACKNOWLEDGMENTS These studies were supported by grants from the National Institute on Aging (AG-07892 to D.G.M. and AG-07909 to C.E.F.); the Anna Greenwall Award from the American Federation for Aging Research; the American Heart Association (891079) to D.G.M.; and the John D. and Katherine T. MacArthur Foundation to C.E.F. D.G.M. is an Established Investigator of the American Heart Association. J.R.G. was supported by a predoctoral fellowship from the National Institute on Aging (AG-00093).
REFERENCES 1. CHANG, H-C., T. SEKI, T. MORIUCIII, AND J. SILVER. 1985. Isolation and characterization of mouse Thy-l genomic clones. Proc. Nod. Acad. Sci. USA 82: 3619-3823.
2. FINCH, C. E., J. R. FOSTER, AND A. E. MIRSKY. 1969. Ageing and the regulation of cell activities during exposure to cold. J. Gen. Physiol. 6: 690-712. 3. FINCH, C. E., AND F. G. GIRGIS. 1974. Enlarged seminal vesicles of senescent C57B1/6J mice. J. Gerontol. 29: 134-138. 4. FINCH, C., S. JOHNSON, S. KOHAMA, S. LERNER, J. MASTERS, P. MAY, D. MORGAN, N. NICHOLS, G. PASINJXTI, AND N. TELFORD. 1987. Physiological approaches to the roles of gene regulation in the brain during aging. In Banbwy Report 27: Molecular Neuropathology of Aging (P. Davis and C. E. Finch, Eds.), pp. 143-158. Cold Springs Harbor Laboratory, Cold Spring Harbor, NY. 5. GOSS, J. R., K-M. CHAN, AND D. G. MORGAN. 1989. Glial fibrillary acidic protein mRNA increases with age in the mouse hippocampus and cerebellum. Sot. Neurosci. Abstr. 19: 106.4. 6. Goss, J. R., C. E. FINCH, AND D. G. MORGAN. 1990. Age-related changes in glial fibrillary acidic protein RNA in the mouse brain. Submitted for publication. 7. JOHNSON, S. A., D. G. MORGAN, AND C. E. FINCH. 1986. Extensive postmortem stability of RNA from rat and human brain. J. Neurosci Res. 16: 267-280. 8. KANG, J., H-G. LEMAIRE, A. UNTEREECK, J. M. SALBAUM, C. L. MASTERS, K-H. GRZESCHIK, G. MULTHAUP, K. BEYREUTHER, AND B. MULLER-HILL. 1987. The precursor of Alzheimer’s disease amyloid A4 protein resembles a cell-surface receptor. Nature (London) 326: 733-736. 9. LEWIS, S. A., J. M. BALCAREK, V. KREK, M. SHELANSKI, AND N. J. COWAN. 1984. Sequence of a cDNA clone encoding mouse glial fibrillary acidic protein: Structural conservation of intermediate filaments. Proc. Natl. Acad. Sci. USA 81: 2743-2746. 10. O’FARRELL, P. H. 1975. High resolution two-dimensional trophoresis of proteins. J. Biol. Chem. 250: 4007-4021.
elec-
11. SHEN, L-P., R. L. PICTET, AND W. L. RU~TER. 1982. Human somatostatin. I. Sequence of the cDNA. Proc. Natl. Acad. Sci. USA 79:4575-4579. 12. VALENZUELA, P., M. QUIROGA, J. ZALDIVAR, W. J. RUTTER, M. W. KIRSHNER, AND D. W. CLEVELAND. 1981. Nucleotide and corresponding amino acid sequence encoded by a and p tubulin mRNAs. Nature (London) 289: 650-656.
108,266-268
(1990)
BRIEF COMMUNICATION GFAP RNA increases during a Wasting State in Old Mice JAMES R. GOSS,~ CALEB E. FINCH, Andrus
Gerontology
Center
and Department
of Biological
Sciences,
University
A prolonged wasting condition is often associated with morbidity in older humans. The effects that such a state has on the quality/quantity of RNA are not known. In initial attempts to develop an animal model for premortem wasting, we examined whole brain RNA from mice slowly approaching death from natural causes. Glial fibrillary acidic protein (GFAP) RNA showed a three-fold increase as detected by RNA gel-blot hybridization analysis. Five other RNA sequences were stable under these circumstances. We conclude that brain RNA changes are selective during a degenerating premortem state. Moreover, RNA sequence changes in conditions such as Alzheimer’s disease should be considered in the context of the wasting condition of the individual and may not be due to a direct effect of the disease process. 0 1990 Academic Press, Inc.
In studying chronic neurodegenerative diseases that, like Alzheimer’s disease (AD), arise later in life, a crucial issue is the impact of chronic wasting conditions. AD patients in the terminal phases often are cachectic, anorexic, immobile, and frequently die from oxygen insufficiency secondary to pulmonary infection. Animal models for a prolonged wasting condition are needed to assesssuch influences on brain RNA integrity. Therefore, we examined the brains of mice that were slowly approaching death from natural causesduring aging. We compared total RNA yields, in vitro translation efficiencies, two-dimensional gel profiles, and selected RNA sequences.The only change observed was a threefold increase in the prevalence of RNA coding for GFAP. Female C57BL/GNNia (Charles River; Wilmington, MA) and male C57BL/6J (Jackson Laboratories; Bar Harbor, ME) mice were used. Animals were housed three to five per cage in an aging colony within the Gerontology Center Vivarium. They were maintained on a 12-h/12-h light/dark cycle with food and water available.
1 To whom
0014~4666/90
all correspondence
should
$3.00
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
be addressed. 266
AND
DAVID G. MORGAN
of Southern
California,
Los Angeles,
California
90089-0191
Approximately 70 mice, aged 24 to 35 months, were observed daily for 4 months. Within this time 15% showed signs of a wasting condition and were used for this study. Wasting mice typically failed to ambulate when gently prodded, exhibited slight resting tremors, displayed a hunched posture, and often felt cold to the touch. After 2 consecutive days of these symptoms, mice were killed by cervical dislocation followed by decapitation; whole brain was removed and stored at -70°C. Agematched mice of the same cohort that did not display any of the symptoms were used as controls. A total of seven pairs of animals were examined. Necropsies were performed on all mice. The wasting mice typically revealed gross abnormalities such as tumors of the mesenteric lymph nodes, enlarged spleens, discolored livers, and empty stomachs. Wasting mice typically weighed less than their control mice (26.6 + 2 vs 29.0 + 1.7 g). These abnormalities and weight loss are consistent with previous findings (2, 3). Control mice were healthy in all regards. Whole brain total RNA was extracted using the guanidine thiocyanate and cesium chloride centrifugation method. Eight micrograms of each RNA sample was electrophoresed through a 1% agarose gel containing 6% formaldehyde and transferred to nylon membranes. Membranes were hybridized to one of six cRNA probes. Riboprobes were transcribed from cDNAs inserted into the transcription vector Bluescribe (Stratagene, San Diego, CA) and labeled with [32P]UTP to a specific activity of 10’ cpm/pg. Clones used were as follows: human @amyloid, 1 kb in length corresponding to nt 1794-2871 of the amyloid precursor gene (8); chicken @-tubulin, 1.2 kb in length corresponding to nt 475-1617 (12); mouse GFAP, 1.1 kb in length corresponding to the entire coding region (9); feline glutamic acid decarboxylase (GAD), 1.25 kb in length corresponding to the amino terminal coding region; human somatostatin (SS), 0.45 kb in length corresponding to part of the preprohormone coding region (11); and mouse Thy-l antigen, 0.76 kb in length corresponding to the entire coding region (1). Blots were hybridized overnight at 77°C in 5X SSC containing 0.5% nonfat dry milk, 1% SDS, 10% dextran sulfate, 25 pg/ml poly(A) RNA, 25 pg/ml poly(C) RNA, and
BRIEF
A
Condition
B
c
GFAP -W
267
COMMUNICATION
5-
WHWHWHWH
THY-1
GFAP
GAD
THY-1
B-TUB
88
B-AMY
RNA Sequence8 FIGS. 1A AND 1B. RNA gel-blot hybridizations to glial fibrillary acidic protein and Thy-l antigen. Eight micrograms of mouse whole brain total RNA was loaded into each lane of a 1% agarose, 6% formaldehyde gel. RNA was transferred to a nylon membrane. The immobilized RNA was hybridized with a 32P-labeled cRNA antisense probe to GFAP (A) or Thy-l antigen (B) at 77°C overnight. The blots were washed in successively lower concentrations of SSC and autoradiographs were taken. These blots represent four of the seven pairs of samples. W, wasting mouse; H, healthy age-matched control mouse. FIG. 1C. Effect of wasting state on selected RNA populations. Autoradiographs were analyzed by computer-aided densitometry. The optical densities (OD) of matched pairs of wasting and healthy mice were compared. A ratio of 1 corresponds to no difference between wasting and healthy mice. GFAP, glial fibrillary acidic protein; GAD, glutamic acid decarboxylase; Thy-l, Thy-l antigen; B-Tub, fl-tubulin; SS, Somatostatin; B-Amy, @-amyloid. P < 0.01, Wilcoxon-Mann-Whitney two-sample test.
100 pg/ml sheared salmon sperm DNA. Blots were washed a total of five times with decreasing concentrations of SSC to a final criterion of 0.5X SSC at 77°C. Kodak X-OMAT film was used for autoradiographs. Autoradiographs were analyzed using a computer-aided densitometry program. Four micrograms of total RNA from each sample was translated in vitro using a rabbit reticulocyte lysate system (BRL). A lo-p1 reaction contained 87 mA4 K+, 1 mM Mg+ and used [35S]methionine to label translation products. Radioactive amino acid incorporation into proteins was measured by placing 2 ~1of the reaction mixture into 1 ml of 1 N NaOH/1.5% Hz02 incubating for 10 min at 37°C followed by TCA precipitation. An amount of the translation mixture equal to approximately 100,000 cpm was used for two-dimensional gel analysis (10). Whole brain RNA yields were similar for wasting and control mice (440 + 40 vs 470 -t 80 pg total RNA/g tissue). Ethidium bromide stains of electrophoresed total RNA showed no evidence of RNA degradation. The radioactive amino acid incorporation in the in vitro translation assay was similar for both wasting and control mice (12,600 + 2400 vs 11,700 + 1000 cpm, respectively). Radioactive translation products were resolved on twodimensional gels and showed no qualitative differences in the 200 major spots (data not shown). Northern blot analysis using the six cRNA probes described above showed a large increase in the signal for GFAP RNA (Fig. 1A). There was no change in RNA for GAD, Thy-l (Fig. lB), fl-tubulin, SS, or P-amyloid. Au-
toradiographs were analyzed by computer-aided densitometry and optical density (OD) measurements were made for all samples hybridized to all cRNA probes. A ratio between wasting and control matched pairs was calculated. A three-fold increase was found for GFAP RNA (Fig. 1C). RNA degradation as assessedfrom these Northern blots was minimal. This study shows an increase in the RNA level for GFAP in response to a wasting condition in mice. This increase is above the already elevated GFAP RNA prevalence found in healthy aged mice (5, 6). None of the other five mRNAs were changed. The wasting condition had no effect on the quantity or quality of whole brain total RNA, as measured by total yields, Northern blot band tailing, or in vitro translation efficiencies and products. In AD we found a similar increase in GFAP in frontal and temporal cortex (4). We originally interpreted this as a response to the neurodegeneration associated with the disease but now consider it possible that this increase in GFAP RNA is a response to the premorbid condition associated with AD, rather than a direct effect of the disease process. This suggests that the premortem condition of the tissue donor should be considered when interpreting gene expression differences found in postmortem human brain tissue. In reference to AD, it should be noted that the wasting condition of mice did not affect the amount of P-amyloid precursor protein RNA. However since this study was conducted using whole brain RNA, we cannot rule out small, regionally
268
BRIEF
COMMUNICATION
localized changes in specific messages. Similarly, changes in RNA levels do not necessarily indicate changes in the corresponding protein. To our knowledge, this is the first study describing changes in RNA associated with the premortem state of an animal. We feel that this condition is a variable which should be considered when examining any changes involving gene expression. Furthermore, since many of the wasting animals had visible abnormalities which may influence results, we emphasize the need for necropsies of aged animals in all applicable studies when possible. While it is presently not possible to identify whether any one abnormality visible at necropsy is responsible for the increase in GFAP RNA, we have observed elevated GFAP RNA in one other study which correlated with grossly observable tumors apparent at necropsy (Goss, Finch, and Morgan, unpublished data). Other extraneous variables that may theoretically influence RNA integrity include postmortem interval, medication history, and nutrition. In previous studies, we found no influence of postmortem interval on the quantity or quality of brain RNA up to 48 h in young rats (7). GFAP was not examined in this study. ACKNOWLEDGMENTS These studies were supported by grants from the National Institute on Aging (AG-07892 to D.G.M. and AG-07909 to C.E.F.); the Anna Greenwall Award from the American Federation for Aging Research; the American Heart Association (891079) to D.G.M.; and the John D. and Katherine T. MacArthur Foundation to C.E.F. D.G.M. is an Established Investigator of the American Heart Association. J.R.G. was supported by a predoctoral fellowship from the National Institute on Aging (AG-00093).
REFERENCES 1. CHANG, H-C., T. SEKI, T. MORIUCIII, AND J. SILVER. 1985. Isolation and characterization of mouse Thy-l genomic clones. Proc. Nod. Acad. Sci. USA 82: 3619-3823.
2. FINCH, C. E., J. R. FOSTER, AND A. E. MIRSKY. 1969. Ageing and the regulation of cell activities during exposure to cold. J. Gen. Physiol. 6: 690-712. 3. FINCH, C. E., AND F. G. GIRGIS. 1974. Enlarged seminal vesicles of senescent C57B1/6J mice. J. Gerontol. 29: 134-138. 4. FINCH, C., S. JOHNSON, S. KOHAMA, S. LERNER, J. MASTERS, P. MAY, D. MORGAN, N. NICHOLS, G. PASINJXTI, AND N. TELFORD. 1987. Physiological approaches to the roles of gene regulation in the brain during aging. In Banbwy Report 27: Molecular Neuropathology of Aging (P. Davis and C. E. Finch, Eds.), pp. 143-158. Cold Springs Harbor Laboratory, Cold Spring Harbor, NY. 5. GOSS, J. R., K-M. CHAN, AND D. G. MORGAN. 1989. Glial fibrillary acidic protein mRNA increases with age in the mouse hippocampus and cerebellum. Sot. Neurosci. Abstr. 19: 106.4. 6. Goss, J. R., C. E. FINCH, AND D. G. MORGAN. 1990. Age-related changes in glial fibrillary acidic protein RNA in the mouse brain. Submitted for publication. 7. JOHNSON, S. A., D. G. MORGAN, AND C. E. FINCH. 1986. Extensive postmortem stability of RNA from rat and human brain. J. Neurosci Res. 16: 267-280. 8. KANG, J., H-G. LEMAIRE, A. UNTEREECK, J. M. SALBAUM, C. L. MASTERS, K-H. GRZESCHIK, G. MULTHAUP, K. BEYREUTHER, AND B. MULLER-HILL. 1987. The precursor of Alzheimer’s disease amyloid A4 protein resembles a cell-surface receptor. Nature (London) 326: 733-736. 9. LEWIS, S. A., J. M. BALCAREK, V. KREK, M. SHELANSKI, AND N. J. COWAN. 1984. Sequence of a cDNA clone encoding mouse glial fibrillary acidic protein: Structural conservation of intermediate filaments. Proc. Natl. Acad. Sci. USA 81: 2743-2746. 10. O’FARRELL, P. H. 1975. High resolution two-dimensional trophoresis of proteins. J. Biol. Chem. 250: 4007-4021.
elec-
11. SHEN, L-P., R. L. PICTET, AND W. L. RU~TER. 1982. Human somatostatin. I. Sequence of the cDNA. Proc. Natl. Acad. Sci. USA 79:4575-4579. 12. VALENZUELA, P., M. QUIROGA, J. ZALDIVAR, W. J. RUTTER, M. W. KIRSHNER, AND D. W. CLEVELAND. 1981. Nucleotide and corresponding amino acid sequence encoded by a and p tubulin mRNAs. Nature (London) 289: 650-656.