N. Leigh Anderson

Selected Bibliography

GENERAL REVIEWS

1. Back to the future: the human protein index (HPI) and the agenda for post-proteomic biology. Anderson, N. G., Matheson, A., and Anderson, N. L. Proteomics 1(1), 3-12. (2001).

The effort to produce an index of all human proteins (the human protein index, or HPI) began twenty years ago, before the initiation of the human genome program. Because DNA sequencing technology is inherently simpler and more scalable than protein analytical technology, and because the finiteness of genomes invited a spirit of rapid conquest, the notion of genome sequencing has displaced that of protein databases in the minds of most molecular biologists for the last decade. However, now that the human genome sequence is nearing completion, a major realignment is under way that brings proteins back to the center of biological thinking. Using an influx of new and improved protein technologies--from mass spectrometry to re-engineered two-dimensional (2-D) gel systems, the original objectives of the HPI have been expanded and the time frame for its execution radically shortened. Several additional large scale technology efforts flowing from the HPI are also described.

2. Proteomics: applications in basic and applied biology. Anderson, N. L., Matheson, A. D., and Steiner, S. Curr Opin Biotechnol 11(4), 408-12. (2000).

The rapid evolution of proteomics has continued during the past year, with a series of innovations in the core technologies of two-dimensional electrophoresis and mass spectrometry, and a diversity of productive research programmes. Well-annotated proteomics databases are now emerging in a number of fields to provide a platform for systematic research, with particularly promising progress in clinical applications such as cardiology and oncology. Large-scale quantitative research, comparable in power and sensitivity to that achieved for gene expression, is thus becoming a reality at the protein level.

3. Proteome and proteomics: new technologies, new concepts, and new words. Anderson, N. L. and Anderson, N. G. Electrophoresis 19(11), 1853-61. (1998).

The goal of proteomics is a comprehensive, quantitative description of protein expression and its changes under the influence of biological perturbations such as disease or drug treatment. Quantitative analysis of protein expression data obtained by high-throughput methods has led us to define the concept of "regulatory homology" and use it to begin to elucidate the basic structure of gene expression control in vivo. Such investigations lay the groundwork for construction of comprehensive databases of mechanisms (cataloguing possible biological outcomes), the next logical step after the soon to be completed cataloguing of genes and gene products. Mechanism databases provide a roadmap towards effective therapeutic intervention that is more direct than that offered by conventional genomics approaches.

4. Twenty years of two-dimensional electrophoresis: past, present and future. Anderson, N. G. and Anderson, N. L. Electrophoresis 17(3), 443-53. (1996).

The history of 2DE is reviewed against the background of changing paradigms in biomedical research, and of the gradual acceptance of two-dimensional analytical techniques in other fields. Future problems concern how 2DE of proteins will interface with, and benefit from, the avalanche of new genomic information now being generated, and how the entire process will be automated. In the long term, a major use for 2DE will be to follow global changes in gene expression during development, in response to drugs and toxic agents, in normal aging, and in various pathological states. Alternative methods have been and are being developed for some of the uses previously thought unique to 2DE, such as the discovery of tissue specific proteins. However, for pharmacological studies involving large sets of proteins, 2DE remains the method of choice for detecting small quantitative changes in abundance in each of a large number of proteins. It is proposed that pharmacological, toxicological and drug discovery studies will comprise major future uses of 2DE, and that these studies will generate the requirement for large scale automation.

PLASMA PROTEOME

1. Analysis of changes in acute-phase plasma proteins in an acute inflammatory response and in rheumatoid arthritis using two-dimensional gel electrophoresis. Doherty, N. S., Littman, B. H., Reilly, K., Swindell, A. C., Buss, J. M., and Anderson, N. L. Electrophoresis 19(2), 355-63. (1998).

Two-dimensional (2-D) gel analysis was used to examine differences in the levels of 19 plasma proteins: before and after an acute inflammatory reaction (parenteral typhoid vaccination) in normal subjects, between rheumatoid arthritis (RA) patients and normals and in RA patients treated with tenidap (120 mg) and piroxicam (20 mg). Typhoid vaccination increased levels of SAA, haptoglobin alpha1, haptoglobin alpha2, haptoglobin beta and alpha1-anti-chymotrypsin but decreased transthyretin and apolipoprotein E. In RA patients, serum amyloid A (SAA), haptoglobin alpha2, haptoglobin beta, alpha1-antichymotrypsin and C3 proactivator levels were elevated while apolipoprotein A-I, apolipoprotein A-IV, transthyretin, Gc-globulin, alpha2-HS glycoprotein, alpha2-macroglobulin and alpha1-B glycoprotein levels were decreased, compared to normals. Compared to piroxicam, tenidap lowered levels of alpha1-antiprotease and SAA but raised the levels of transthyretin, Gc-globulin, alpha2-HS-glycoprotein and alpha2-macroglobulin in RA patients. C-reactive protein (CRP) could not be quantified on 2-D gels but, when measured by rate nephelometry, levels were reduced after treatment with tenidap compared to piroxicam. The general pattern of the acute phase protein response to an acute inflammatory response to typhoid vaccination is similar to that in the chronic inflammatory condition, RA. The impact of tenidap on both positive and negative acute-phase proteins in RA patients could clearly be distinguished from that of piroxicam.

2. A two-dimensional gel database of human plasma proteins. Anderson, N. L. and Anderson, N. G. Electrophoresis 12(11), 883-906. (1991).

An updated two-dimensional electrophoretic map of human plasma proteins is presented, together with a complete listing of the individual protein spots, their locations, size and isoelectric points relative to internal charge standards. Forty-nine polypeptide species are identified, many consisting of multiple spots differing in glycosylation or sequence (e.g., immunoglobulins). A further series of 35 as yet uncharacterized proteins is indicated.

3. Use of high-resolution two-dimensional gel electrophoresis for analysis of monoclonal antibodies and their specific antigens. Pearson, T. W. and Anderson, N. L. Methods Enzymol 92, 196-220. (1983).

4. Microheterogeneity of serum transferrin, haptoglobin and alpha 2 HS glycoprotein examined by high resolution two-dimensional electrophoresis. Anderson, N. L. and Anderson, N. G. Biochem Biophys Res Commun 88(1), 258-65. (1979).

5. High resolution two-dimensional electrophoresis of human plasma proteins. Anderson, L. and Anderson, N. G. Proc Natl Acad Sci U S A 74(12), 5421-5. (1977).

The two-dimensional electrophoretic technique of O'Farrell has been adapted to the analysis of human plasma proteins, and 30 polypeptides have been identified in the pattern produced. Genetic variants involving charge (isoelectric point) or size (molecular weight in the presence of sodium dodecyl sulfate) changes should be routinely detectable in at least 20 proteins at once, facilitating studies of human mutation rates.

6. High resolution two-dimensional electrophoretic mapping of immunoglobulin light chains. Anderson NL. Immunol Lett 2, 195-199. (1981).

OTHER BODY FLUIDS

1. Proteins of human urine. I. Concentration and analysis by two-dimensional electrophoresis. Anderson, N. G., Anderson, N. L., and Tollaksen, S. L. Clin Chem 25(7), 1199-210. (1979).

BLOOD CELLS

1. Analysis of human leukemic cells by use of high-resolution two-dimensional electrophoresis. I: results of a pilot study. Anderson, N. L., Wiltsie, J. C., Li, C. Y., Willard-Gallo, K. E., Tracy, R. P., Young, D. S., Powers, M. T., and Anderson, N. G. Clin Chem 29(5), 762-7. (1983).

We analyzed mononuclear leukocytes from patients with various human leukemias by high-resolution two-dimensional electrophoresis. Tumor cells of the granulocytic, monocytic, and lymphoid lineages [obtained from chronic granulocytic leukemia in blast transformation, acute monocytic leukemia, and chronic lymphocytic leukemia (CLL), respectively] can be easily recognized by using a series of cell-type marker proteins identified by comparison of fractionated normal cell populations. B and T cell types of CLL could be distinguished, the results correlating well with those obtained by use of monoclonal-antibody staining methods. In two cases representing almost pure B-cells (classical CLL; 0% T, 85% B) and T-cells (cutaneous T-cell leukemia; 77% T, 0% B), 27 of 29 marker proteins showed quantitative B/T differences comparable to those observed in comparisons of normal B-and T-lymphocytes prepared by cell sorting. These results indicate that cells from relatively well-differentiated leukemias show complex patterns of gene expression very similar to those of the corresponding normal cells and strongly support the use of large marker panels in cell-type determination. Less-well-differentiated acute leukemias [such as acute undifferentiated and acute granulocytic (FAB:M1)] appear to yield protein patterns corresponding less closely to recognizable mature cell types, and may show expression of novel proteins related to the state of differentiation.

2. Lymphocyte, monocyte, and granulocyte proteins compared by use of two-dimensional electrophoresis. Gemmell, M. A. and Anderson, N. L. Clin Chem 28(4 Pt 2), 1062-6. (1982).

We compared cellular proteins from normal human blood lymphocytes, monocytes and granulocytes, using high-resolution two-dimensional electrophoresis. The leukocytes were isolated from peripheral blood by centrifugation on density step gradients (yielding a fraction of purified granulocytes and a lymphocyte/monocyte mixture) and monocytes were subsequently separated from lymphocytes by virtue of their adherence to plastic. Wright-stained smears indicated that each of the three resulting fractions was 90 to 95% pure. The cells were labeled with [35S]methionine after various intervals in culture, then solubilized and analyzed by two-dimensional electrophoresis. Although most proteins in each cell type are common to all three, there are nevertheless several specific marker proteins that distinguish one cell type from another. We also examined the appearance of these markers in three lines of cultured cells from humans (GM607, a B-lymphoblastoid line; HL-60, a promyelocytic leukemic line; and 1494, a normal skin fibroblast).

3. Red cell proteins. I. Two-dimensional mapping of human erythrocyte lysate proteins. Edwards, J. J., Anderson, N. G., Nance, S. L., and Anderson, N. L. Blood 53(6), 1121-32. (1979).

Human erythrocyte lysate proteins were resolved into over 250 discrete spots by two-dimensional electrophoresis using isoelectric focusing in the first dimension and electrophoresis in the presence of sodium dodecyl sulfate, (SDS) in the second. The overwhelming excess of hemoglobin has made such analyses difficult in the past. However, with the ISO-DALT two-dimensional electrophoresis system, large numbers of red cell proteins can be mapped in the presence of hemoglobin. When hemoglobin and several other major proteins are removed by adsorption to DEAE-cellulose, additional minor components are seen, giving a total of over 275. With the use of purified preparations, the map positions of five cell enzymes or their subunits were determined: pyruvate kinase, catalase, glucose-6-phosphate dehydrogenase, hypoxanthine phosphoribosyltransferase, and carbonic anhydrase. The mapping techniques described complement and extend those traditionally used to find human red cell protein variants.

HUMAN PROTEIN INDEX

1. Design and implementation of a prototype Human Protein Index. Taylor, J., Anderson, N. L., Scandora, A. E. Jr., Willard, K. E., and Anderson, N. G. Clin Chem 28(4 Pt 2), 861-6. (1982).

This paper describes information-handling aspects of the TYCHO I analysis system (Clin, Chem. 27: 1807--1820, 1981), which analyzes two-dimensional electrophoresis gels, matches the individual protein spots with those in a reference pattern, and stores various information--including spot measurements, identifications, treatment profiles, set memberships, and comments--in a computerized database. This and additional information such as amino acid composition and cellular localization is then accessible from an interactive program that includes a pictorial user interface and presents much of the data in graphical form. Use of the TYCHO I system is illustrated by examples drawn from analyses of gel patterns from human leukocytes.

2. Photo/essay. The human protein index. Anderson, N. G. and Anderson, N. L. Jama 246(22), 2620-1 . (1981).

3. Muscle protein analysis. I. High-resolution two-dimensional electrophoresis of skeletal muscle proteins for analysis of small biopsy samples. Giometti, C. S., Anderson, N. G., and Anderson, N. L. Clin Chem 25(11), 1877-84. (1979).

We have been developing a clinically useful method for high-resolution two-dimensional electrophoretic analysis of small (5--10 mg) human muscle biopsy samples with sufficient resolution to resolve the major contractile proteins and enzymes. Using rabbit psoas muscle as a model, we describe methods for sample preparation and two-dimensional electrophoresis. Basic proteins, which appear as streaks when conventional isoelectric focusing is used in the first dimension, are resolved through a modification of the nonequilibrium pH gradient electrophoresis method [Cell 12, 1133 (1977)]. In the two-dimensional patterns obtained from rabbit muscle, we identify the components of 10 enzymes and of myosin, actin, tropomyosin, and troponin. These patterns indicate charge heterogeneity in a large fraction of the proteins. Comparison of rabbit and normal human muscle patterns shows many similarities, but much additional work is required to confirm identifications. We conclude that analysis of small biopsy samples is feasible, but that all aspects of human sample acquisition, storage (when necessary), and preparation require thorough study before the method becomes routine in human muscle research and, ultimately, in the diagnosis of some muscle diseases.

RODENT STUDIES

1. Cholesterol biosynthesis regulation and protein changes in rat liver following treatment with fluvastatin. Steiner, S., Gatlin, C. L., Lennon, J. J., McGrath, A. M., Seonarain, M. D., Makusky, A. J., Aponte, A. M., Esquer-Blasco, R., and Anderson, N. L. Toxicol Lett 120(1-3), 369-77. (2000).

The enzyme 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase is a key regulator in cholesterol biosynthesis and HMG CoA reductase inhibitors (statins) have become a widely prescribed family of lipid lowering agents. Cholesterol synthesis occurs predominantly in liver which is the target organ of statins. We studied the effects of fluvastatin (Lescol), a member of the statin family, on hepatic protein regulation. Male F344 rats treated with 0.8 mg/kg per day fluvastatin or 24 mg/kg per day fluvastatin for 7 days showed treatment-related changes in 58 liver proteins (P<0.005). Major effects were evident in the cholesterol biosynthesis pathway including the induction of enzymes upstream and downstream of the target enzyme HMG CoA reductase. Treatment also triggered alterations in key enzymes of carbohydrate metabolism and was associated with changes in a heterogeneous set of cellular stress proteins involved in cytoskeletal structure, calcium homeostasis and protease activity. The latter set of protein alterations indicates that hepatotoxicity is associated with high-dose treatment. Based on the results it is suggested that HMG-CoA synthase and isopentenyl-diphosphate delta-isomerase may be explored as alternative drug targets and that the induction levels of these enzymes may serve as a measure of potency of individual statin drugs. It is proposed that efficacy and cellular stress markers discovered in this study may be used in a high throughput screen (HTS) assay format to compare efficiently and accurately the therapeutic windows of different members of the statin family.

2. Pharmaceutical proteomics. Steiner, S. and Anderson, N. L. Ann N Y Acad Sci 919, 48-51. (2000).

Genomics and proteomics are today well established in drug discovery and, in combination with combinatorial chemistry and high-throughput screening, are helping to bring forward an unprecedented number of potential lead compounds. To avoid the generation of bottlenecks downstream in drug development, increasing pressure is arising to integrate these technologies into the development environment. Proteomics has demonstrated proof-of-concept in toxicology as shown by a number of successful applications in mechanistic toxicology and lead selection. The "technology wave" is now starting to impact the clinical phase of drug development. Expected benefits are optimized clinical trials based on the availability of biologically relevant markers of drug efficacy and safety.

3. Expression profiling in toxicology--potentials and limitations. Steiner, S. and Anderson, N. L. Toxicol Lett 112-113, 467-71. (2000).

Recent progress in genomics and proteomics technologies has created a unique opportunity to significantly impact the pharmaceutical drug development processes. The perception that cells and whole organisms express specific inducible responses to stimuli such as drug treatment implies that unique expression patterns, molecular fingerprints, indicative of a drug's efficacy and potential toxicity are accessible. The integration into state-of-the-art toxicology of assays allowing one to profile treatment-related changes in gene expression patterns promises new insights into mechanisms of drug action and toxicity. The benefits will be improved lead selection, and optimized monitoring of drug efficacy and safety in pre-clinical and clinical studies based on biologically relevant tissue and surrogate markers.

4. The effects of peroxisome proliferators on protein abundances in mouse liver. Anderson, N. L., Esquer-Blasco, R., Richardson, F., Foxworthy, P., and Eacho, P. Toxicol Appl Pharmacol 137(1), 75-89. (1996).

We have investigated the effects of five peroxisome proliferators (PPs : clofibric acid, DEHP, WY14,643, nafenopin, and LY171883) on the abundances of a large number of proteins in the livers of treated mice at 5- and 35-day time points. LY171883 was investigated at a range of doses, and one of its close structural analogs that is not a peroxisome proliferator (LY163443) was included as a negative control compound. Liver samples were analyzed by quantitative 2-D electrophoresis. Data for a selected set of 107 liver protein spots that respond strongly to at least one of the test compounds was subjected to principal component analysis to search for global protein pattern changes. The first component (PC1) accounted for 51% of the total data variance and was identified as a global measure of peroxisome proliferation by its correlation with enzymatic peroxisomal beta-oxidation. Component PC2 (7%) separated 5- and 35-day exposures, and PC3 (5%) separated groups treated with LY163443 from the rest. We used PC1 as a surrogate for equivalent dose in order to examine the effects of diverse compounds, with widely differing potencies, on a common scale. Analyzed in this way, the data indicate that all the peroxisome proliferators tested produce effects over wide time and dose ranges that fall on or near a single curve. Examination of specific protein responses showed that many proteins individually show a unified response curve, but that curves for different proteins were different. In particular, it appears that some constitutive proteins showing modest inductions with a high dose plateau (such as cytosolic epoxide hydrolase) are inducible at lower doses than some proteins showing very strong, nonplateaued inductions (such as the 80-kDa peroxisomal bifunctional enzyme). The results provide support for a unified receptor-based mechanism controlling the main PP response, but demonstrate that individual responsive genes can show quite different dose-response curves.

5. Cyclosporine A decreases the protein level of the calcium-binding protein calbindin-D 28kDa in rat kidney. Steiner, S., Aicher, L., Raymackers, J., Meheus, L., Esquer-Blasco, R., Anderson, N. L., and Cordier, A. Biochem Pharmacol 51(3), 253-8. (1996).

Despite the widespread use of cyclosporine A (CsA), its mechanism of action and side effects are not yet completely understood. There exists a large body of evidence suggesting that disturbance of calcium homeostasis is a critical step in the cascade of cellular and molecular events induced by the drug. As recently shown in our laboratory by two-dimensional protein gel electrophoresis (2-DE) analysis of kidney homogenates, CsA induced numerous changes in several kidney proteins. One kidney protein in particular was shown to be strongly down-regulated by the drug. In this work we report the identification of the strongly decreased kidney protein as calbindin-D 28kDa, a vitamin D-dependent calcium-binding protein associated with calcium handling by cells. The assignment of the down-regulated protein spot is based on its internal amino acid sequence analysis and its specific reaction with a monoclonal antibody raised against calbindin-D 28kDa. In kidney homogenates of male Wistar rats treated with 50 mg/kg/d CsA for up to 28 days, calbindin levels were measured by ELISA and were shown to be continuously decreased with prolonged CsA treatment. To our knowledge, this is the first report describing the effect of CsA on kidney calbindin-D 28kDa protein levels. Further studies are needed to elucidate whether the CsA-mediated down-regulation of the calcium-binding protein calbindin-D 28kDa may be a critical factor for the renal adverse effects induced by this drug.

6. An updated two-dimensional gel database of rat liver proteins useful in gene regulation and drug effect studies. Anderson, N. L., Esquer-Blasco, R., Hofmann, J. P., Meheus, L., Raymackers, J., Steiner, S. , Witzmann, F., and Anderson, N. G. Electrophoresis 16(10), 1977-81. (1995).

We have improved upon the reference two-dimensional (2-D) electrophoretic map of rat liver proteins originally published in 1991 (N. L. Anderson et al., Electrophoresis 1991, 12, 907-930). A total of 53 proteins (102 spots) are now identified, many by microsequencing. In most cases, spots cut from wet, Coomassie Blue stained 2-D gels were submitted to internal tryptic digestion [2], and individual peptides, separated by high-performance liquid chromatography (HPLC), were sequenced using a Perkin-Elmer 477A sequenator. Additional spots were identified using specific antibodies.

7. The hepatocarcinogen methapyrilene but not the analog pyrilamine induces sustained hepatocellular replication and protein alterations in F344 rats in a 13-week feed study. Cunningham, M. L., Pippin, L. L., Anderson, N. L., and Wenk, M. L. Toxicol Appl Pharmacol 131(2), 216-23. (1995).

Methapyrilene (MPH) was a widely used antihistamine until it was found to produce hepatocellular carcinoma and cholangiocarcinoma in Fischer 344 rats. The structurally similar antihistamine pyrilamine (PYR) was marginally or noncarcinogenic in a similar study. The peroxisome proliferator Wy-14,643 was included in this study as a positive control. As part of a program to investigate the mechanisms whereby structurally similar chemicals produce different toxicities, we studied these three chemicals for the induction of cell proliferation in the liver of F344 rats. Male rats were treated for up to 13 weeks with feed dosed with MPH (HCl salt) at 0, 50, 100, 250, or 1000 ppm or PYR (maleate salt) at 1000 ppm to duplicate the route of administration and high-dose groups used in the carcinogenesis assay. In addition, the nongenotoxic hepatocarcinogen peroxisome proliferator Wy-14,643 was included as a positive cell-proliferating chemical. Cell proliferation was quantitated by measuring the incorporation of bromodeoxyuridine (BrDU) administered by osmotic minipump for 7 days and the appearance of proliferating cell nuclear antigen (PCNA) immunohistochemically. The BrDU-labeling index showed a large and sustained increase in rats treated with MPH at 250 and 1000 ppm, sustaining greater than 50% labeling in the higher dose group of 4-, 6-, and 13-week treatment groups. PYR at 1000 ppm demonstrated no significant increase in labeling above control levels at any time point. PCNA-labeling indexes showed similar but reduced increases for MPH and were comparable to control for the PYR dose groups. Two-dimensional gel electrophoresis was used for the detection of quantitative changes in gene expression and qualitative changes in the charges of specific mitochondrial and cytosolic proteins. Quantitative changes in 32 proteins induced by MPH and 39 changes induced by Wy-14,643 were detected throughout the 13-week study. Specific mitochondrial protein charge shifts were associated with high-dose MPH treatment that were not observed in animals treated with Wy-14,643. PYR induced no significant qualitative or quantitative protein alterations. Hepatocellular proliferation of the large magnitude observed following dietary administration of MPH, and not PYR may contribute to the mechanism of carcinogenesis of MPH.

8. A comparative study of mouse liver proteins arylated by reactive metabolites of acetaminophen and its nonhepatotoxic regioisomer, 3'-hydroxyacetanilide. Myers, T. G., Dietz, E. C., Anderson, N. L., Khairallah, E. A., Cohen, S. D., and Nelson, S. D. Chem Res Toxicol 8(3), 403-13. (1995).

Acetaminophen (4'-hydroxyacetanilide), a widely used analgesic/antipyretic drug, is hepatotoxic in large doses, whereas the m-hydroxy isomer of acetaminophen, 3'-hydroxyacetanilide, is not hepatotoxic. Both are oxidized by mouse liver cytochromes P-450 to reactive metabolites that bind covalently to hepatic proteins. Because previous studies have shown that peak levels of liver protein adducts formed after the administration of each of these compounds to mice are nearly equivalent, and because liver protein adduct formation correlates with hepatotoxicity caused by acetaminophen in mice, we investigated the abundance and patterns of protein adducts formed by acetaminophen and its regioisomer for significant differences. Hepatotoxic doses of acetaminophen to mice significantly altered the abundances of several liver proteins 2 h after dosing as revealed by densitometric analysis of two-dimensional electrophoretic patterns of these proteins. The same analysis after the administration to mice of 3'-hydroxyacetanilide indicated that this nonhepatotoxic regioisomer of acetaminophen caused several similar changes in protein patterns, but also revealed some significant differences. Binding of radiolabeled acetaminophen and 3'-hydroxyacetanilide to hepatic proteins corroborated and extended these results. Two hours after the administration of 14C-labeled analogs of these two compounds to mice, at a time when their extent of total covalent binding to hepatic proteins is approximately equivalent, there are many similarities but also some differences in selectivity of proteins that are adducted, as revealed by both one-dimensional and two-dimensional gel electrophoresis followed by phosphorimage analysis of radiolabel bound to protein bands. Moreover, protein adducts formed from 3'-hydroxyacetanilide were found to be less stable than those formed from acetaminophen under the conditions of electrophoretic analysis. Furthermore, a comparison of radiodetection and immunodetection of protein adducts formed from acetaminophen with an antibody specific for acetaminophen protein adducts indicates that the antibody detects most of the same proteins that are radiolabeled and that the relative quantitative contribution of various adducts to the overall covalent binding of acetaminophen to proteins is approximately the same by both methods. Thus, 3'-hydroxyacetanilide should prove to be a useful tool to aid in the discrimination of hepatic acetaminophen protein adducts that may be critical or noncritical to survival of hepatocytes.

9. Effects of toxic agents at the protein level: quantitative measurement of 213 mouse liver proteins following xenobiotic treatment. Anderson, N. L., Giere, F. A., Nance, S. L., Gemmell, M. A., Tollaksen, S. L., and Anderson, N. G. Fundam Appl Toxicol 8(1), 39-50. (1987).

By analyzing two-dimensional electrophoretic patterns of mouse liver proteins with a computerized image analysis system, we have observed quantitative changes in the abundance of more than 70 proteins in mice treated with various agents. Aroclor 1254, a mixture of polychlorinated biphenyls known to induce a broad spectrum of microsomal activity, induces the largest group of changes (60 proteins altered at p less than 0.001 significance). Phenobarbital produces a small set of characteristic changes that forms part of the much larger Aroclor 1254 effect. Ibuprofen treatment produces a phenobarbital-like pattern of change, with the addition of at least one protein change not observed with any of the other treatments. Cycloheximide and carbon tetrachloride each induces a different characteristic pattern of protein alteration. We have assigned most of the mouse liver proteins to a specific subcellular fraction, and it appears that the predominant class of proteins altered by each compound is present in the soluble phase, rather than in the microsomal fraction. The ability to survey large numbers of tissue proteins for involvement in pharmacologic and toxic effects may allow a more comprehensive understanding of the mechanisms of action in vivo and provide new markers of tissue damage.

10. Proteomic analysis of rat soleus and tibialis anterior muscle following immobilization. Isfort, R. J., Wang, F., Greis, K. D., Sun, Y., Keough, T. W., Bodine, S. C., and Anderson, N. L. J Chromatogr B Analyt Technol Biomed Life Sci 769(2), 323-32. (2002).

A proteomic analysis was performed comparing normal slow twitch type fiber rat soleus muscle and normal fast twitch type fiber tibialis anterior muscle to immobilized soleus and tibialis anterior muscles at 0.5, 1, 2, 4, 6, 8 and 10 days post immobilization. Muscle mass measurements demonstrate mass changes throughout the period of immobilization. Proteomic analysis of normal and atrophied soleus muscle demonstrated statistically significant changes in the relative levels of 17 proteins. Proteomic analysis of normal and atrophied tibialis anterior muscle demonstrated statistically significant changes in the relative levels of 45 proteins. Protein identification using mass spectrometry was attempted for all differentially regulated proteins from both soleus and tibialis anterior muscles. Four differentially regulated soleus proteins and six differentially regulated tibialis anterior proteins were identified. The identified proteins can be grouped according to function as metabolic proteins, chaperone proteins, and contractile apparatus proteins. Together these data demonstrate that coordinated temporally regulated changes in the proteome occur during immobilization-induced atrophy in both slow twitch and fast twitch fiber type skeletal muscle.

11. Proteomic analysis of rat soleus muscle undergoing hindlimb suspension-induced atrophy and reweighting hypertrophy. Isfort, R. J., Wang, F., Greis, K. D., Sun, Y., Keough, T. W., Farrar, R. P., Bodine, S. C., and Anderson, N. L. Proteomics 2(5), 543-50. (2002).

A proteomic analysis was performed comparing normal rat soleus muscle to soleus muscle that had undergone either 0.5, 1, 2, 4, 7, 10 and 14 days of hindlimb suspension-induced atrophy or hindlimb suspension-induced atrophied soleus muscle that had undergone 1 hour, 8 hour, 1 day, 2 day, 4 day and 7 days of reweighting-induced hypertrophy. Muscle mass measurements demonstrated continual loss of soleus mass occurred throughout the 21 days of hindlimb suspension; following reweighting, atrophied soleus muscle mass increased dramatically between 8 hours and 1 day post reweighting. Proteomic analysis of normal and atrophied soleus muscle demonstrated statistically significant changes in the relative levels of 29 soleus proteins. Reweighting following atrophy demonstrated statistically significant changes in the relative levels of 15 soleus proteins. Protein identification using mass spectrometry was attempted for all differentially regulated proteins from both atrophied and hypertrophied soleus muscle. Five differentially regulated proteins from the hindlimb suspended atrophied soleus muscle were identified while five proteins were identified in the reweighting-induced hypertrophied soleus muscles. The identified proteins could be generally grouped together as metabolic proteins, chaperone proteins and contractile apparatus proteins. Together these data demonstrate that coordinated temporally regulated changes in the skeletal muscle proteome occur during disuse-induced soleus muscle atrophy and reweighting hypertrophy.

12. Proteomic analysis of the atrophying rat soleus muscle following denervation. Isfort, R. J., Hinkle, R. T., Jones, M. B., Wang, F., Greis, K. D., Sun, Y., Keough, T. W., Anderson, N. L., and Sheldon, R. J. Electrophoresis 21(11), 2228-34. (2000).

A proteomic analysis was performed comparing normal rat soleus muscle to denervated soleus muscle at 0.5, 1, 2, 4, 6, 8 and 10 days post denervation. Muscle mass measurements demonstrated that the times of major mass changes occurred between 2 and 4 days post denervation. Proteomic analysis of the denervated soleus muscle during the atrophy process demonstrated statistically significant (at the p < 0.01 level) changes in 73 soleus proteins, including coordinated changes in select groups of proteins. Sequence analysis of ten differentially regulated proteins identified metabolic proteins, chaperone and contractile apparatus proteins. Together these data indicate that coordinated temporally regulated changes in the proteome occur during denervation-induced soleus muscle atrophy, including changes in muscle metabolism and contractile apparatus proteins.

13. Proteomics to display lovastatin-induced protein and pathway regulation in rat liver. Steiner, S., Gatlin, C. L., Lennon, J. J., McGrath, A. M., Aponte, A. M., Makusky, A. J., Rohrs, M. C., and Anderson, N. L. Electrophoresis 21(11), 2129-37. (2000).

Lovastatin is a lipid lowering agent that acts by inhibiting 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase, a key regulatory enzyme in cholesterol biosynthesis. In this study the pattern of gene network regulation induced in hepatic proteins as a response to lovastatin treatment was analyzed by proteomics. In livers of male F344 rats treated with 1.6 mg/kg/day lovastatin or 150 mg/kg/day lovastatin for seven days, 36 proteins were found to be significantly altered (p<0.001) in relation to treatment. The changed proteins were classified according to their cellular function and participation in biochemical pathways. The following observations were made: (i) inhibition of HMG-CoA reductase provoked a regulatory response in the cholesterol synthesis pathway including the induction of cytosolic HMG-CoA synthase and of isopentenyl-diphosphate delta-isomerase, (ii) manipulation of the lipid metabolism triggered alterations in key enzymes of the carbohydrate metabolism, and (iii) lovastatin treatment was associated with signs of toxicity as reflected by changes in a heterogeneous set of cellular stress proteins involved in functions such as cytoskeletal structure, calcium homeostasis, protease inhibition, cell signaling or apoptosis. These results present new insights into liver gene network regulations induced by lovastatin and illustrate a yet unexplored application of proteomics to discover new targets by analysis of existing drugs and the pathways that they regulate.

14. Two-dimensional electrophoresis of liver proteins: characterization of a drug-induced hepatomegaly in rats. Newsholme, S. J., Maleeff, B. F., Steiner, S., Anderson, N. L., and Schwartz, L. W. Electrophoresis 21(11), 2122-8. (2000).

Two-dimensional electrophoresis (2-DE) of liver proteins was applied to further characterize an unusual drug-induced increase in hepatocellular rough endoplasmic reticulum (RER) in Sprague-Dawley rats given a substituted pyrimidine derivative. Absolute liver weights of drug-treated rats (9.9 +/- 0.4 g) increased above vehicle-treated controls (7.2 +/- 0.2 g) by 37%. Light microscopy revealed diffuse granular basophilia of the hepatocellular cytoplasm, uncharacteristic of hepatocytes and suggested cells rich in ribosomes, which was confirmed by electron microscopy. Immunostaining for cell proliferation, viz., 5-bromo-2'-deoxyuridine (BrdU) and proliferating cell nuclear antigen (PCNA), indicated marked hepatocellular proliferative activity. 2-DE of solubilized liver using an ISO-DALT gel system indicated significant (p<0.001) quantitative changes in at least 17 liver proteins (12 increased, 5 decreased) compared to controls. The protein with the largest increase was homologous to acute-phase reactant, contrapsin-like protein inhibitor-6. Other markedly upregulated proteins were methionine adenosyltransferase, a catalyst in methionine/ATP metabolism and mitochondrial HMG-CoA synthase, involved in cholesterol synthesis. The complementary strategies of 2-DE coupled either with database spot mapping or protein isolation and amino acid sequencing successfully identified a subset of proteins from xenobiotic-damaged rodent livers, the expression of which differed from controls. However, the current bioinformatics platform for rodent hepatic proteins and limited knowledge of specific protein functionality restricted application of this proteomics profile to further define a mechanistic basis for this unusual hepatotoxicity.

15. Confirmed identities of proteins from a two-dimensional map of Syrian hamster embryo cells. Asquith, T. N., Gauggel, D. L., Esquer-Blasco, R., Anderson, N. L., and Isfort, R. J. Electrophoresis 20(7), 1646-51. (1999).

The Syrian hamster embryo (SHE) cell transformation assay is widely used to screen chemicals for carcinogenic potential. However, the biochemical mechanisms of transformation in SHE cells are incompletely understood relative to other rodent systems. Thus identification of proteins which change during transformation can provide clues to biochemical mechanisms. Previously, we published a map of SHE cell proteins based on comparisons to other maps. In this report we provide direct sequence analysis of numerous proteins which were previously identified solely by electrophoretic mobility. Protein sequencing verified original spot identifications and extended the range of identified proteins. The updated map will assist in evaluating biochemical mechanisms of morphological transformation in hamster cells.

16. Cholangiocyte-specific rat liver proteins identified by establishment of a two-dimensional gel protein database. Tietz, P., de Groen, P. C., Anderson, N. L., Sims, C., Esquer-Blasco, R., Meheus, L., Raymackers, J., Dauwe, M., and LaRusso, N. F. Electrophoresis 19(18), 3207-12. (1998).

The liver is composed of a variety of cells that form a functional unit involved in uptake, synthesis, metabolism, and secretion. Until recently, most studies examining liver function did not analyze the specific proteins expressed or functions performed by the multiple individual cell types that constitute the hepatic mass. In the last decade, novel isolation methods have been developed that allow the purification of liver cell populations highly enriched in one type of liver cell. Here, we present a detailed two-dimensional (2-D) protein map of rat bile duct epithelial cells (i.e., cholangiocytes) using a recently developed isolation procedure. In addition, we identify 27 major cholangiocyte proteins either by comparison to maps of known rat liver proteins (based on pI and Mr) or by tryptic digestion and microsequencing. Finally, we compare the relative abundance of individual proteins present in cholangiocytes to whole liver as well as hepatocyte-specific proteins. Our results show that cholangiocytes express a unique array of individual proteins. The cholangiocyte 2-D protein pattern is markedly different from that of isolated rat hepatocytes or whole rat liver, with high levels of proteins previously known to be expressed by cholangiocytes (e.g., cytokeratins, actins) as well as protein not previously demonstrated to be expressed at high levels (e.g., annexin V, selenium binding protein). We conclude that this cholangiocyte-derived, 2-D protein map will be a crucial resource for studies directed at our understanding of cholangiocyte physiology and pathobiology.

17. Changes in the liver protein pattern of female Wistar rats treated with the hypoglycemic agent SDZ PGU 693. Arce, A., Aicher, L., Wahl, D., Anderson, N. L., Meheus, L., Raymackers, J., Cordier, A., and Steiner, S. Life Sci 63(25), 2243-50. (1998).

SDZ PGU 693 acts as a hypoglycemic agent by stimulating glucose utilisation in insulin-sensitive peripheral tissues, such as skeletal muscle and fat. In a 28 day toxicity study the compound was found to induce hepatocellular hypertrophy in Wistar rats treated with 300 mg/kg/day. To gain insights into the pathomechanism of these alterations, aliquots of liver samples from control and treated female Wistar rats were separated by two-dimensional protein gel electrophoresis and the digitized images of the protein patterns were searched for protein abundance changes. Significant treatment-related quantitative changes (P < 0.001) were found in 29 liver proteins. Major increases were observed in several microsomal proteins, including NADPH cytochrome P-450 reductase, cytochrome b5 and serine protease inhibitor. The changes in the cytochrome related enzymes, both known co-factors of the P-450 enzyme system, strongly suggest that SDZ PGU 693 induces microsomal proliferation and induction of the P-450 enzyme system. Decreases were observed in a series of mitochondrial proteins, such as F1ATPase-delta subunit and ornithine aminotransferase precursor as well as in several cytosolic proteins such as the liver fatty acid binding protein, arylsulfotransferase and the senescence marker protein-30. The changes in F1ATPase-delta subunit and liver fatty acid binding protein together suggest a down-regulation of the mitochondrial liver fatty acid metabolism, likely reflecting the pharmacological action of the compound. These results show that SDZ PGU 693 produces a complex pattern of gene expression changes which give insights into the molecular mechanisms of both its pharmacological action and a toxic response.

18. Induction of the adipose differentiation-related protein in liver of etomoxir-treated rats. Steiner, S., Wahl, D., Mangold, B. L., Robison, R., Raymackers, J., Meheus, L., Anderson, N. L., and Cordier, A. Biochem Biophys Res Commun 218(3), 777-82. (1996).

The effects of etomoxir, an irreversible carnitine palmitoyltransferase I inhibitor, on the liver protein pattern and on liver morphology were examined by two-dimensional gel electrophoresis in female Sprague-Dawley rats treated with 125 mg/kg/day etomoxir for 28 days. In livers of treated animals a protein spot was found which was not present in controls. The spot was identified by internal amino acid sequence analysis as the adipose differentiation-related protein (ADRP). The expression of ADRP in liver is a novel finding as the protein has been described previously as adipocyte-specific. Additionally we found histopathologic evidence of lipid accumulation in the livers of etomoxir rats. The data show that for each treated rat there was a good correlation between ADRP levels and degree of lipid droplet formation. This observation may suggest a potential relationship between drug-induced expression of ADRP in liver and lipid accumulation.

19. Simultaneous measurement of hundreds of liver proteins: application in assessment of liver function. Anderson, N. L., Taylor, J., Hofmann, J. P., Esquer-Blasco, R., Swift, S., and Anderson, N. G. Toxicol Pathol 24(1), 72-6. (1996).

Proteins implement most biological functions at the molecular level. As one might expect based on this fact, it appears that the altered functional states associated with toxic effects involve changes in the abundance or structure of proteins. Although numerous specific assays exist to measure changes in the abundance of individual proteins, practical limitations have prevented widespread use of multiple protein assays for the global characterization of toxicity. Recent developments in protein analytical technology are rapidly changing this picture. Two-dimensional gel electrophoresis, a technique capable of resolving and quantitating hundreds of proteins simultaneously, is becoming an automated, high-throughput tool. In parallel, techniques have been developed that allow the resulting deluge of protein measurements to be organized into a prototype Molecular Effects Database describing xenobiotic effects in rodent liver. This database can detect, classify, and characterize a broad range of liver toxicity mechanisms. It currently contains approximately 10 million protein measurements, including data on the liver effects of 43 compounds, with a further 50 compounds to be added in 1995. Observed effects range from very broad (sex steroids alter levels of 45% of all liver proteins) to very specific (e.g., hepatic hydroxymethyl glutaryl coenzyme A reductase inhibitors). Companion 2-dimensional databases describing rodent brain and kidney have been initiated, as have linkages to the genomic sequence databases. Assimilation of this approach into research and regulatory toxicology poses an interesting challenge--one that is likely to lead to a radically more sophisticated understanding of toxicity and its biological basis.

20. Dose-responses in rat hepatic protein modification and expression following exposure to the rat hepatocarcinogen methapyrilene. Richardson, F. C., Horn, D. M., and Anderson, N. L. Carcinogenesis 15(2), 325-9. (1994).

Dose-related effects of methapyrilene (MP) on protein modification and expression were examined using two-dimensional gel electrophoresis (2-D PAGE) coupled with computer analysis. Methapyrilene was administered ad libitum at doses of 0, 62.5, 125, 250 and 1000 p.p.m. to male F-344 rats for 12 weeks beginning at 8 weeks of age. Following treatment, livers were removed and frozen for 2-D PAGE analysis. Separation of hepatic proteins was conducted using ISO-DALT technology. Changes in abundance and modification of hepatic proteins were determined using the Kepler software package. Covalent modifications of three specific mitochondrial proteins were quantified using a charge modification index. Dose-response relationships were analyzed using Tukey's trend test. Results demonstrated that covalent modification of the three mitochondrial proteins was linearly related to dose and that a dose effect could be found at all dose levels in 2 out of 3 proteins. Two forms of change in protein expression were observed: (i) a dose-dependent change with effects at all doses and (ii) a change only at the toxic dose of 1000 p.p.m. MP. These results demonstrate a molecular effect of MP at doses that do not produce cellular responses including toxicity or increases in cell replication suggesting that these specific mitochondrial modifications are molecular dosimeters but are probably not direct factors and/or sufficient factors in carcinogenesis. This study also demonstrates the potential use of 2-D PAGE electrophoresis to delineate the effect of dose on expression of specific proteins.

21. Comparisons of protein changes in human and rodent hepatocytes induced by the rat-specific carcinogen, methapyrilene. Richardson, F. C., Strom, S. C., Copple, D. M., Bendele, R. A., Probst, G. S., and Anderson, N. L. Electrophoresis 14(1-2), 157-61. (1993).

There is a growing concern that the rodent bioassay may not always serve as an appropriate model to assess the carcinogenic risk for humans exposed to certain compounds. Mechanistic research that examines the effects of a compound in rodent and man could help in the interpretation of bioassay results. This paper reports a novel use of two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) technology to assess similarities and differences in the response of rodents and humans to the rat-specific hepatocarcinogen, methapyrilene (MP). A sequential examination of rodent and human hepatic proteins was conducted following in vivo exposure of rats and mice and in vitro exposure of rat, mouse, and human hepatocytes to MP. Results showed that covalent modifications observed in rats and mice in vivo were duplicated both qualitatively and quantitatively in the corresponding in vitro systems and that these modifications correlated with carcinogenic susceptibility. Covalent modifications in human hepatocytes were minimal despite exposure to concentrations of MP that were 6-fold higher than those used in rodent hepatocytes. These studies suggest that in the case of MP the rat is not the most appropriate model for assessing the human situation. Furthermore, these data show that in vitro-in vivo comparisons based on 2-D PAGE may provide adjunctive information for extrapolating rodent toxicity/bioassay results to human risk assessment.

22. Two-dimensional gel electrophoresis analysis of Syrian hamster embryo cells: morphological transformation is not cell type specific. Isfort, R. J., Kerckaert, G., Anderson, N. L., and LeBoeuf, R. A. Electrophoresis 13(11), 855-61. (1992).

Studies were conducted to investigate the protein phenotype of normal and morphologically transformed Syrian hamster embryo (SHE) cells. Based upon two-dimensional gel protein phenotype analysis, we conclude that (i) SHE cells are a mixture of multiple cell types including mesenchymal and epithelial cells and (ii) several cell types present in the SHE cell population can be morphologically transformed by a variety of genotoxic and non-genotoxic carcinogens.

23. Covalent protein modifications and gene expression changes in rodent liver following administration of methapyrilene: a study using two-dimensional electrophoresis. Anderson, N. L., Copple, D. C., Bendele, R. A., Probst, G. S., and Richardson, F. C. Fundam Appl Toxicol 18(4), 570-80. (1992).

The effect of methapyrilene (MP), a mitochondrial proliferator and presumed nongenotoxic carcinogen, has been examined in rodent liver by means of high-resolution two-dimensional electrophoretic analysis of total proteins. Using this approach, we have discovered protein modifications in rat liver resulting from 1 week MP treatment that suggest the involvement of a reactive drug metabolite. The restriction of these molecular charge modifications to mitochondrial proteins indicates that such a reactive metabolite must be generated and confined within the mitochondrion. Quantitative changes in numerous nonmitochondrial proteins are also observed. Following a 4-week recovery period, almost all the 1-week treatment changes are reversed, reestablishing a protein pattern close to that of the controls. At the end of a 10-week exposure, the mitochondrial protein modifications are increased and are accompanied by a variety of quantitative protein changes indicative of a large shift in gene expression and/or cell type composition. When a 4-week untreated recovery period follows the 10-week treatment, small quantitative changes persist. In the mouse, where MP appears not to induce mitochondrial proliferation or tumorigenesis, 1 week treatment nevertheless produces mitochondrial protein changes in vivo consistent with attack by a reactive metabolite, but at a level substantially lower than that seen in the rat. Features of the mitochondrial protein modification indicate that it is covalent, does not involve cysteine or tryptophan, and results from binding of a negatively charged adduct. The possibility that the putative reactive metabolite could also attack mitochondrial (but not nuclear) DNA suggests that MP could be genotoxic in an unconventional way. Detection of protein modification by two-dimensional gel analysis appears to offer a general method for the detection and characterization of processes generating reactive metabolites.

24. A two-dimensional gel database of rat liver proteins useful in gene regulation and drug effects studies. Anderson, N. L., Esquer-Blasco, R., Hofmann, J. P., and Anderson, N. G. Electrophoresis 12(11), 907-30. (1991).

A standard two-dimensional (2-D) protein map of Fischer 344 rat liver (F344MST3) is presented, with a tabular listing of more than 1200 protein species. Sodium dodecyl sulfate (SDS) molecular mass and isoelectric point have been established, based on positions of numerous internal standards. This map has been used to connect and compare hundreds of 2-D gels of rat liver samples from a variety of studies, and forms the nucleus of an expanding database describing rat liver proteins and their regulation by various drugs and toxic agents. An example of such a study, involving regulation of cholesterol synthesis by cholesterol-lowering drugs and a high-cholesterol diet, is presented. Since the map has been obtained with a widely used and highly reproducible 2-D gel system (the Iso-Dalt system), it can be directly related to an expanding body of work in other laboratories.

CELL LINES

1. Specific protein phosphorylation in human promyelocytic HL-60 leukemia cells susceptible or resistant to induction of cell differentiation by phorbol-12-myristate-13-acetate. Anderson, N. L., Gemmell, M. A., Coussens, P. M., Murao, S., and Huberman, E. Cancer Res 45(10), 4955-62. (1985).

The pattern of protein phosphorylation induced by phorbol-12-myristate-13-acetate (PMA) was analyzed by two-dimensional gel electrophoresis in human HL-60 leukemia cells, which are susceptible to induction of cell differentiation by PMA, and in cells from an HL-60 cell variant designated R-94 that are resistant to such an induction. Protein phosphorylation was detected by observing either a rapid acid-directed charge shift of [35S]methionine-labeled protein or an increase in the amount of phosphate label in a 32P-labeled protein. The results indicated that PMA at 10(-7) M causes within 30 min after treatment the phosphorylation of at least ten different proteins in both the HL-60 and R-94 cells. Among these ten phosphorylated proteins, we identified a major cytoplasmic polypeptide (Mr approximately 64,000), a cytoskeletal protein (Mr approximately 56,000), a nonmuscle myosin light chain, and two proteins (Mr approximately 60,000 and 64,000) localized in or around the cell nucleus. Phosphoamino acid analysis of six of the ten phosphoproteins showed that they contain phosphoserine. None of these proteins contained phosphotyrosine or phosphothreonine. The R-94 cell variant was found to be capable of increased protein phosphorylation after PMA treatment; however, the level of phosphate incorporation reached only the level of the untreated HL-60 cells and thus fell far short of the level observed in the HL-60 cells after PMA treatment. It is suggested that the basis for the acquired resistance in R-94 cells towards induction of cell differentiation by PMA is a block in signal transmission involving phosphorylation of nuclear protein(s) following the binding of the inducer PMA to its receptor (protein kinase C).

2. Protein-pattern changes and morphological effects due to methionine starvation or treatment with 5-azacytidine of the phorbol-ester-sensitive cell lines HL-60, CCL-119, and U-937. Anderson, N. L. and Gemmell, M. A. Clin Chem 30(12 Pt 1), 1956-64. (1984).

Methionine starvation causes changes in the protein pattern of HL-60 promyelocytic leukemia cells as observed by two-dimensional electrophoresis. One group of proteins is apparently modified, appearing in new positions. A further series of proteins, including several principal nuclear polypeptides, is substantially diminished. The morphology of a fraction of the cells in the culture changes concomitantly, with condensation and fragmentation of the nucleus and eventual remolding of the cell to a "grape-cluster" appearance. Similar effects are produced by a DNA methylation inhibitor, 5-azacytidine, but not by various other toxic agents tested. A defect in DNA methylation, either by depletion of S-adenosyl-L-methionine (the methyl donor) or by inactivation of the relevant enzyme, may be responsible. The T-lymphoblastoid line CCL-119 and the histiocytic lymphoma line U-937 also show these effects, but most fibroblast, epithelial, and lymphoblastoid lines do not. These changes can be largely prevented in each of the three susceptible lines by prior treatment with the tumor promoter, phorbol myristate acetate (PMA), an agent known to cause differentiation in at least two of the lines. The results thus suggest interesting relationships between methionine metabolism, protein and structural changes in the cell nucleus, and PMA-induced cell differentiation.

3. Protein variants in human cells: enumeration by protein indexing. Anderson, N. L., Giometti, C. S., Gemmell, M. A., and Macy, M. Ann N Y Acad Sci 428, 134-43. (1984).

4. Control of macrophage cell differentiation in human promyelocytic HL-60 leukemia cells by 1,25-dihydroxyvitamin D3 and phorbol-12-myristate-13-acetate. Murao, S., Gemmell, M. A., Callaham, M. F., Anderson, N. L., and Huberman, E. Cancer Res 43(10), 4989-96. (1983).

Human promyelocytic leukemia cells (HL-60) were induced to differentiate into macrophage-like cells in a dose (3 X 10(-10) to 10(-7) M) and time (1 to 6 days)-dependent manner by 1,25-dihydroxyvitamin D3 and the tumor promoter, phorbol-12-myristate-13-acetate. Differentiation was determined by an increase in the percentage of morphologically mature cells, in lysozyme and nonspecific esterase activities, and in reactivity with the murine OKM1 monoclonal antibody. Two HL-60 cell variants, designated as R-80 and B-II, were also examined. R-80 cells, which are resistant to induction of cell differentiation by phorbol-12-myristate-13-acetate, also exhibited resistance, although to a lesser degree, to induction of cell differentiation by 1,25-dihydroxyvitamin D3. The resistance to the action of the two compounds is presumably not due to similar binding sites for the two inducers, since 1,25-dihydroxyvitamin D3 was unable to compete for the phorbol diester binding sites as measured by [3H]phorbol-12,13-dibutyrate binding. B-II cells were resistant to induction of cell differentiation by 1,25-dihydroxyvitamin D3, phorbol-12-myristate-13-acetate, retinoic acid, and dimethyl sulfoxide. Two-dimensional electrophoretic analysis of HL-60 cell protein patterns indicated that treatment of the HL-60 cells with 1,25-dihydroxyvitamin D3, phorbol-12-myristate-13-acetate, retinoic acid, and dimethyl sulfoxide caused the cells to express various monocyte-macrophage and granulocyte marker proteins. None of the inducers caused a protein pattern identical to that of peripheral monocytes or granulocytes in the HL-60 cells, but the protein pattern of the HL-60 cells treated with 1,25-dihydroxyvitamin D3 was the closest to that of peripheral blood monocytes. These results indicate that 1,25-dihydroxyvitamin D3 induces in the HL-60 cells a phenotype that resembles, but is not identical to, that of peripheral monocytes-macrophages.

5. Cytoskeletal proteins from human skin fibroblasts, peripheral blood leukocytes, and a lymphoblastoid cell line compared by two-dimensional gel electrophoresis. Giometti, C. S., Willard, K. E., and Anderson, N. L. Clin Chem 28(4 Pt 2), 955-61. (1982).

Differences in proteins between cells grown as suspension cultures and those grown as attached cultures were studied by comparing the proteins of detergent-resistant cytoskeletons prepared from peripheral blood leukocytes and a lymphoblastoid cell line (GM607) (both grown as suspension cultures) and those of human skin fibroblasts (grown as attached cultures) by two-dimensional gel electrophoresis. The major cytoskeletal proteins of the leukocytes were also present in the protein pattern of GM607 cytoskeletons. In contrast, the fibroblast cytoskeletal protein pattern contained four groups of proteins that differed from the patterns of the leukocytes and GM607. Three groups (Cytoskf:8--10, :14--16, and :17--18) showed qualitative differences, and the fourth group (Cytoskf:11 and :13) showed quantitative differences. In addition, surface labeling of GM607 and human fibroblasts with 125I demonstrated that substantial amounts of vimentin and actin are exposed at the surface of the attached fibroblasts, but there is little evidence of similar exposure at the surface of the suspension-grown GM607. Cytoskf:11 and :13 in fibroblast preparations were also labeled with the 125I. These results demonstrate some differences in cytoskeletal protein composition between different types of cells could be related to their ability or lack of ability to grow as attached cells in tissue culture.

6. A two-dimensional electrophoretic analysis of the heat-shock-induced proteins of human cells. Anderson, N. L., Giometti, C. S., Gemmell, M. A., Nance, S. L., and Anderson, N. G. Clin Chem 28(4 Pt 2), 1084-92. (1982).

Using two-dimensional electrophoresis, we have investigated the responses of human cells in culture to heat shock and to various chemical agents producing a similar effect. These treatments result in the induction of increased synthesis of several specific proteins. One (HShock:1, SDS-molecular mass about 65000) is increased by about 350-fold over the amount in untreated cells. Computer analysis of time-course studies indicates, however, that rates of synthesis of various proteins other than the classical "heat shock proteins" are affected, some of these alterations following time courses quite different from the main (HShock) inductions. The heat shock effect is thus much more complicated than previously realized. We purified the HShock:1 protein from heat-shocked human lymphoblastoid cells, and prepared a rabbit antiserum specific for HShock:1 on nitrocellulose two-dimensional gel transfers of total lymphoblastoid cell protein. A survey of mouse tissues shows high concentrations of an HShock:1-like protein in the testis, and human testes also appears to contain substantial (though lower) concentrations. These results are consistent with the hypothesis (derived from the tissue-culture studies) that the heat shock effect is a general response to the need for increased protein catabolism within the cell. Increased concentrations of HShock:1 are also observed in preparations of blood leukocytes collected from patients after surgery, indicating that some types of physiological trauma may induce the heat shock proteins in man. Using the antiHShock:1 antibody in an immunoassay, it will be possible to systematically examine HShock:1 concentrations in plasma and leukocytes, thereby opening up the possibility of a clinical test based for the first time upon an inducible aspect of cellular gene expression.

7. Two different variants of the same tropomyosin polypeptide in clones from GM1386 human skin fibroblasts. Giometti, C. S., Gemmell, M. A., and Anderson, N. L. Biochem Biophys Res Commun 128(3), 1247-53. (1985).

A new protein observed in two-dimensional electrophoresis patterns of proteins from the human skin fibroblast line GM1386 has been identified as a charge and molecular-weight variant of the type of tropomyosin found in smooth muscle (Tm:3). This is the second variant of Tm:3 found in GM1386 and represents a second site mutation in one of the genes coding for Tm:3.

8. Tropomyosin heterogeneity in human cells. Giometti, C. S. and Anderson, N. L. J Biol Chem 259(22), 14113-20. (1984).

Tropomyosin preparations from human platelets, human peripheral blood leukocytes from normal individuals and from a patient with chronic lymphocytic leukemia, human lymphoblastoid cells (GM607), human epithelial cells, and human skin fibroblasts have all been found to contain more than one protein when analyzed by two-dimensional gel electrophoresis. Although the lymphoid cell preparations consistently contain two proteins of almost identical molecular weight (Mr = 30,000), the platelet, epithelial cell, and fibroblast preparations contain two or more major proteins with molecular weights between 31,000 and 36,000, in addition to a major protein at 30,000. All of these proteins have characteristics in common with tropomyosin including slightly acidic isoelectric point (approximately pH 4), stability to heat and organic solvents, association with the cytoskeleton, and reactivity with antibody against skeletal muscle tropomyosin. The nonmuscle tropomyosin-like proteins were compared with tropomysins from human skeletal, cardiac, and smooth muscle by peptide mapping after partial proteolysis. The results showed one of the non- muscle proteins to be identical to the major smooth muscle tropomyosin in human uterus (myometrium) and another to be similar but not identical to skeletal muscle alpha-tropomyosin. The remainder of the proteins with tropomyosin characteristics was unique to non-muscle cells. In all, nine distinct human proteins with characteristics of tropomyosin are described. Charge variants of two of these proteins have been described previously.

9. An electrophoretic variant of a human fibroblast protein with characteristics of smooth muscle tropomyosin. Giometti, C. S. and Anderson, N. L. J Mol Biol 173(1), 109-23. (1984).

A charge variant of a protein (Tm: 3; molecular weight approximately 35,000) that co-migrates with human smooth muscle tropomyosin has been found in whole cell extracts from the fibroblasts of a father and his son. The variant protein co-purifies with Tm: 3, shifts with it to a higher apparent molecular weight on sodium dodecyl sulfate/polyacrylamide gel electrophoresis in the presence of 4 M-urea, is a component of the microfilaments, and has a peptide cleavage pattern identical to that of Tm: 3 and smooth muscle tropomyosin. The results indicate that Tm: 3 and the variant (Tm: 3.1) are smooth muscle tropomyosin, suggesting that normal human fibroblasts synthesize at least two tropomyosins (Tm:3 and the non-muscle tropomyosin Tm:4, molecular weight approximately 30,000) and that they are the products of separate genes.

10. A variant of human nonmuscle tropomyosin found in fibroblasts by using two-dimensional electrophoresis. Giometti, C. S. and Anderson, N. L. J Biol Chem 256(22), 11840-6. (1981).

In an analysis of 12 human fibroblast cell lines by two-dimensional electrophoresis, one cell line (1493) was found to contain a major protein variant (Cytosk: 12; for the convention used in spot numbering, see Anderson, N. L. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, 2407- 2411) not present in any of the other 11 cell lines. Biochemical characterization of the variant protein included determination of its subcellular location, partial amino acid composition, behavior on sodium dodecyl sulfate (SDS) versus SDS/urea gels, and partial proteolytic digestion patterns. All of these methods showed that Cytosk:12 is related to Cytosk:11. Both proteins are located in the cytoskeleton, contain little cysteine or proline and no detectable tryptophan, shift together to a higher apparent molecular weight when electrophoresed in the presence of SDS and 8 M urea versus SDS alone, and have identical products after partial proteolysis according to Cleveland et al. (Cleveland, D. W., Fischer, S. G., Kirschner, M. W., and Laemmli, U. K. (1977) J. Biol. Chem. 252, 1102-1106). Preparation of nonmuscle tropomyosin resulted in the partial purification of both Cytosk:11 and :12. The data suggest that Cytosk:11 is fibroblast nonmuscle tropomyosin and that Cytosk:12 in cell line 1493 is a charge variant of that protein.

MICROORGANISMS

1. Comparison of African trypanosomes of different antigenic phenotypes, subspecies and life cycle stages by two-dimensional gel electrophoresis. Anderson, N. L., Parish, N. M., Richardson, J. P., and Pearson, T. W. Mol Biochem Parasitol 16(3), 299-314. (1985).

High resolution two-dimensional polyacrylamide gel (2D gel) electrophoresis and autoradiography were used to analyze the protein gene products of African trypanosomes biosynthetically labelled with [35S]methionine. Using cloned populations of parasites it was found that: antigenically different bloodstream trypanosomes from the same serodeme differed only in their variant surface glycoproteins; Trypanosoma brucei, T.b. rhodesiense and T.b. gambiense subspecies could be distinguished on the basis of differences in expressed proteins; transformation from bloodstream trypomastigotes to procyclic epimastigote culture forms was accompanied by loss of variant surface glycoproteins and several other qualitative and quantitative changes in minor proteins. The results indicate that 2D gel analysis may allow improved classification of African trypanosomes (based on the observation of hundreds of protein markers) and may also provide a general technique for the identification of lifecycle stage specific markers.

2. Two-dimensional electrophoresis used to differentiate the causal agents of American tegumentary leishmaniasis. Saravia, N. G., Gemmell, M. A., Nance, S. L., and Anderson, N. L. Clin Chem 30(12 Pt 1), 2048-52. (1984).

American tegumentary leishmaniasis is caused by a diverse group of Leishmania classified within two species complexes: L. mexicana and L. braziliensis. Because distinct disease forms are associated with certain species or subspecies, prognosis requires taxonomic identification of the pathogen. Biological criteria allow identification only to the species level; thus, biochemical and immunological markers are eagerly sought. We have used two-dimensional electrophoresis to examine the relationships among reference strains of New World Leishmania and stocks isolated from Colombian patients. The L. mexicana and L. braziliensis species complexes are shown to be extremely disparate, and the relative affinities of the subspecies reference strains and Colombian Leishmania stocks are documented. The latter displayed a variety of patterns--some clearly identifiable with a particular L. braziliensis subspecies reference strain, others intermediate between those of L.b. panamensis and L.b. guyanensis. Such comparisons are useful both in establishing relatedness and identifying subspecies and variant marker proteins, which may have biological significance.

PROTEOMICS METHODS

1. Analytical techniques for cell fractions. XXIII. A stable thermal gradient device for heat denaturation studies on proteins. Anderson, N. L., Eisler, W. J., and Anderson, N. G. Anal Biochem 91(2), 441-5. (1978).

The ISO-DALT two-dimensional electrophoretic system (1,2), based on the method of O'Farrell (3), is capable of performing large numbers of analysis on complex mixtures of proteins. However, both separations employed are carried out under dissociating or denaturing conditions and no enzyme activities are readily observable in the analyzed proteins. In order to identify the spots corresponding to particular enzymes, it is therefore necessary to employ some nondestructive resolving technique first and as a second step to perform both enzyme and two-dimensional electrophoretic analyses on the fractions generated. By correlating enzyme activity with intensity of various spots on the two-dimensional gels throughout the series of initial fractions, identifications, can be made. This approach, unlike the more direct immunoprecipitation methods (4), requires the running of large numbers of enzyme analyses and two-dimensional gels and some convenient initial resolving procedure. Convenient and rapid techniques for the analyses (5,6) and gels (1,2) have been described previously in this series and elsewhere. This paper deals with the use of selective denaturation in a temperature gradient as an initial resolving procedure and describes a simple thermal gradient device for generating such a gradient.

2. Rapid mass spectrometric identification of proteins from two-dimensional polyacrylamide gels after in gel proteolytic digestion. Li, G., Waltham, M., Anderson, N. L., Unsworth, E., Treston, A., and Weinstein, J. N. Electrophoresis 18(3-4), 391-402. (1997).

We report a rapid method for identifying proteins resolved by two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) using matrix-assisted laser desorption ionization-mass spectrometry (MALDI-MS). In-gel digestion was performed in a way such that the volume ratio of trypsin solution to gel plug was quantitatively controlled to promote reproducible digestion and to maximize the digestion yield. To make the digestion samples more compatible with MALDI-MS, the volatile salt ammonium bicarbonate in the digestion buffer was largely removed prior to peptide extraction. Samples of mixed tryptic peptides from in-gel digestion were used without purification to obtain molecular weights by MALDI-MS with alpha-cyano, 4-hydroxy-cinnamic acid as the matrix. Modifications of MALDI sample loading procedures improved the detection sensitivity by one half to one order of magnitude. The peptide mass peaks in MALDI-MS spectra were distinguished from those of impurities by using several types of controls, and masses were corrected by using trypsin autodigestion fragments as internal calibration standards. Two different peptide-matching computer programs were used to interrogate sequence databases and identify proteins. Identification was enhanced by generation of orthogonal data sets (by using different proteases) and by including experimental values of isoelectric point (pI) and molecular weight to exclude false entries in the candidate lists. Approximately 1% of the material from a spot was used in each sample loading, and nine protein spots from rat liver 2-D PAGE gels were identified correctly, as judged by comparison with identification results previously obtained from Edman sequencing. A previously identified low-abundance spot was not identified by MALDI-MS, presumably because there was insufficient material in a single gel. The sample handling procedure reported here should permit us to identify many 2-D PAGE protein spots of medium abundance.

3. Analytical techniques for cell fractions. XXVII. Use of heart proteins as reference standards in two-dimensional electrophoresis. Giometti, C. S., Anderson, N. G., Tollaksen, S. L., Edwards, J. J., and Anderson, N. L. Anal Biochem 102(1), 47-58. (1980).

4. Analytical techniques for cell fractions. XXVI. A two-dimentional electrophoretic analysis of basic proteins using phosphatidyl choline/urea solubilization. Willard, K. E., Giometti, C. S., Anderson, N. L., O'Connor, T. E., and Anderson, N. G. Anal Biochem 100(2), 289-98. (1979).

5. The beta and cytoplasmic actins are differentially thermostabilized by MgADP; gamma actin binds MgADP more strongly. Anderson, N. L. Biochem Biophys Res Commun 89(2), 486-90. (1979).

6. Analytical techniques for cell fractions. XXV. Concentration and two-dimensional electrophoretic analysis of human urinary proteins. Anderson, N. G., Anderson, N. L., Tollaksen, S. L., Hahn, H., Giere, F., and Edwards, J. Anal Biochem 95(1), 48-61. (1979).

7. Analytical techniques for cell fractions. XXIV. Isoelectric point stadnards for two-dimensional electrophoresis. Anderson, N. L. and Hickman, B. J. Anal Biochem 93(2), 312-20. (1979).

8. Analytical techniques for cell fractions. XXII. Two-dimensional analysis of serum and tissue proteins: multiple gradient-slab gel electrophoresis. Anderson, N. L. and Anderson, N. G. Anal Biochem 85(2), 341-54. (1978).

9. Analytical techniques for cell fractions. XXI. Two-dimensional analysis of serum and tissue proteins: multiple isoelectric focusing. Anderson, N. G. and Anderson, N. L. Anal Biochem 85(2), 331-40. (1978).

10. Analytical techniques for cell fractions. XX. Cyclic affinity chromatography: principles and applications. Anderson, N. G., Willis, D. D., Holladay, D. W., Caton, J. E., Holleman, J. W., Eveleigh, J. W., Attrill, J. E., Ball, F. L., and Anderson, N. L. Anal Biochem 68(2), 371-93. (1975).

11. Analytical techniques for cell fractions. XIX. The cyclum: an automatic system for cyclic chromatography. Anderson, N. G., Willis, D. D., Holladay, D. W., Caton, J. E., Holleman, J. W., Eveleigh, J. W., Attrill, J. E., Ball, F. L., and Anderson, N. L. Anal Biochem 66(1), 159-74. (1975).

12. Anderson, Leigh. Two-dimensional electrophoresis : operation of the ISO-DALT system. (1988). Washington, D.C., Large Scale Biology Press.

BIOINFORMATICS

1. A protein expression database for the molecular pharmacology of cancer. Myers, T. G., Anderson, N. L., Waltham, M., Li, G., Buolamwini, J. K., Scudiero, D. A., Paull, K. D., Sausville, E. A., and Weinstein, J. N. Electrophoresis 18(3-4), 647-53. (1997).

In the last six years, the Developmental Therapeutics Program (DTP) of the US National Cancer Institute (NCI) has screened over 60,000 chemical compounds and a larger number of natural product extracts for their ability to inhibit growth of 60 different cancer cell lines representing different organs of origin. Whereas inhibition of the growth of one cancer cell type gives no information on drug specificity, the relative growth inhibitory activities against 60 different cells constitute patterns that encode detailed information on mechanisms of action and resistance (as reviewed in Boyd and Paull, Drug Devel. Res. 1995, 34, 19-109 and Weinstein et al., Science 1997, 275, 343-349). In order to correlate the patterns of activity with properties of the cells, we and other laboratories are characterizing the cells with respect to a large number of factors at the DNA, mRNA, and protein levels. As part of that effort, we have developed a two-dimensional gel electrophoresis (2-DE) protein expression database covering all 60 cell types (Buolamwini et al., submitted). Here we present analyses of the correlations among protein spots (i) in terms of their patterns of expression and (ii) in terms of their apparent relationships to the pharmacology of a set of 3989 screened compounds. The correlations tend to be stronger for the latter than for the former, suggesting that the spots have more robust signatures in terms of the pharmacology than in terms of expression levels. Links to pertinent databases and tools of analysis will be updated progressively at http:@www.nci.nih.gov/intra/lmp/jnwbio.htm and http:@epnwsl.ncifcrf.gov:2345/dis3d/dtp.++ +html.

2. An information-intensive approach to the molecular pharmacology of cancer. Weinstein, J. N., Myers, T. G., O'Connor, P. M., Friend, S. H., Fornace, A. J. Jr., Kohn, K. W., Fojo, T., Bates, S. E., Rubinstein, L. V., Anderson, N. L., Buolamwini, J. K. , van Osdol, W. W., Monks, A. P., Scudiero, D. A., Sausville, E. A., Zaharevitz, D. W., Bunow, B., Viswanadhan, V. N., Johnson, G. S., Wittes, R. E., and Paull, K. D. Science 275(5298), 343-9. (1997).

Since 1990, the National Cancer Institute (NCI) has screened more than 60,000 compounds against a panel of 60 human cancer cell lines. The 50-percent growth-inhibitory concentration (GI50) for any single cell line is simply an index of cytotoxicity or cytostasis, but the patterns of 60 such GI50 values encode unexpectedly rich, detailed information on mechanisms of drug action and drug resistance. Each compound's pattern is like a fingerprint, essentially unique among the many billions of distinguishable possibilities. These activity patterns are being used in conjunction with molecular structural features of the tested agents to explore the NCI's database of more than 460,000 compounds, and they are providing insight into potential target molecules and modulators of activity in the 60 cell lines. For example, the information is being used to search for candidate anticancer drugs that are not dependent on intact p53 suppressor gene function for their activity. It remains to be seen how effective this information-intensive strategy will be at generating new clinically active agents.

3. Global approaches to quantitative analysis of gene-expression patterns observed by use of two-dimensional gel electrophoresis. Anderson, N. L., Hofmann, J. P., Gemmell, A., and Taylor, J. Clin Chem 30(12 Pt 1), 2031-6. (1984).

A major difficulty in the use of two-dimensional protein maps to identify and classify cell types is the problem of acquiring, selecting, and analyzing quantitative data on hundreds of protein spots. Here we use methods of multivariate statistics to analyze the differences among a panel of human cell lines, in some cases involving quantitative data on more than 250 proteins. Principal-component and cluster-analysis techniques show that the lines can be easily distinguished, even by using the subset of proteins present in all cells. A preliminary analysis of the protein changes brought about by phorbol ester-induced differentiation of the line U937 is included.

4. The TYCHO system for computer analysis of two-dimensional gel electrophoresis patterns. Anderson, N. L., Taylor, J., Scandora, A. E., Coulter, B. P., and Anderson, N. G. Clin Chem 27(11), 1807-20. (1981).

We describe here a computer system for the analysis of high-resolution two-dimensional gel-electrophoresis patterns, with some initial applications. The system (called TYCHO) comprises programs for image acquisition, background subtraction and smoothing, spot detection, gaussian spot modeling, and pattern matching and comparison. It is based on a conventional minicomputer, but makes extensive use of a high-speed array processor in the image-processing and -modeling steps. Used in concert with the ISO-DALT two-dimensional electrophoresis system (Anal. Biochem. 85:331-354, 1978), TYCHO allows quantitative measurement of hundreds of proteins in complex biological samples, and constitutes the initial data-reduction system required for work towards a Human Protein Index.

HEMOGLOBIN

1. Response of the Bohr group salt bridges to ligation of the T state of haemoglobin Kansas. Kilmartin, J. V. and Anderson, N. L. J Mol Biol 123(1), 71-87. (1978).

2. Hemoglobin St. Louis [beta 28 (B10) Leu replaced by Gln]: crystal structure of the fully reduced (deoxy) form. Anderson, N. L. J Clin Invest 58(5), 1107-9. (1976).

The three-dimensional structure of fully reduced Hb St. Louis has been determined to 3.5 A resolution. The difference electron density map clearly shows the site of the mutation and the effects it produces. Glutamine B10 and histidine E7 (the distal histidine) swing towards each other and, between them, stabilize a water molecule in the normally hydrophobic heme pocket. This creation of an aqueous microenvironment near the heme accounts for the thermal instability, high rate of methemoglobin formation, and increased oxygen affinity observed in solution studies of the mutant as described in the preceeding paper. Other than a small increase in tilt of the heme, virtually no further stereochemical disturbances result.

3. Hemoglobin San Diego (beta 109 (G11) val--met). Crystal structure of the deoxy form. Anderson, N. L. J Clin Invest 53(1), 329-33. (1974).

4. Site of the amino-acid substitution in haemoglobin Seattle ( 2 A 2 70 Asp ). Anderson, N. L. and Perutz, M. F. Nat New Biol 243(130), 274-5. (1973).

5. Structures of deoxy and carbonmonoxy haemoglobin Kansas in the deoxy quaternary conformation. Anderson, L. J Mol Biol 94(1), 33-49. (1975).

6. Intermediate structure of normal human haemoglobin: methaemoglobin in the deoxy quaternary conformation. Anderson, L. J Mol Biol 79(3), 495-506. (1973).

OLIGONUCLEOTIDE SYNTHESIS

1. Large-scale oligonucleotide synthesizers. I. Basic principles and system design. Anderson, N. G., Anderson, N. L., Taylor, J., and Goodman, J. Appl Biochem Biotechnol 54(1-3), 19-42. (1995).

The central problem in scaling up oligonucleotide synthesis is to expose each element of a large bed to the same conditions obtaining in very small ones, for the same intervals of time. Our analysis suggests that scale-up is chiefly limited by fluid path length through the bed. By using annular beds in zonal centrifuge rotors of unique design, with fluid flow controlled by combining centrifugal force with differences in physical density between reagents, reagent fronts may be kept exactly perpendicular to the direction of flow in each bed element. Under these conditions, bed volume may be increased by increasing rotor length and diameter. The rotor is lined with polypropylene or Teflon, and has a thick tempered glass end window. Transparent rotary valves of a unique design allow any of 47 reagents to be selected and the direction of flow through the rotor to be controlled. A photodiode spectrophotometer provides complete absorption spectra on fluid in the rotor inlet and outlet lines every 4 s, and a large balance weighs effluent from the synthesizer continuously. The entire operation is controlled by a work station, and steps may be programmed by time, absorbance, or reagent mass. Reagents are identified by spectra, and trityls are integrated on line. A detailed time-stamped log file provides a complete record of each synthesis.