Aromatic Amines , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Alkanes and Alkenes ................................. 48 Tobacco Isoprenoids ................................... 48 Benzenes and Naphthalenes ....................... .49 Polynuclear Aromatic Hydrocarbons (PAH) .. .51 N-Heterocyclic Hydrocarbons (Aza-Arenes) ... .52 Phenols ................................................... 52 Carboxylic Acids ...................................... .57 Metallic Constituents ................................ .58 Radioactive Compounds ............................. .60 Agricultural Chemicals .............................. .61 Tobacco Additives ..................................... 63 Toxic and Carcinogenic Agents-A Summary.. ........ .64 References ......................................................... .66 Physiological Responses to Cigarette Smoke.. ................ .73 Animal Smoke Inhalation Exposure Methodology ..... .73 Smoke Generation ........................................ .73 Methods of Inhalant Delivery.. ...................... .74 Dosimetry ................................................... .74 Limiting Factors in Smoke Exposure.. ............ .75 Selected Animal Studies ...................................... .76 Pulmonary Studies ....................................... .76 Cardiovascular Studies .................................. .76 Exercise Tolerance ........................................ 77 Toxicity of Specific Smoke Components ................. .78 Nicotine ..................................................... .78 Carbon Monoxide ......................................... .79 Nitric Oxide ............................................... .86 Nitrogen Dioxide ......................................... .81 Phenol ....................................................... .81 References .......................................................... 82 Pharmacology of Cigarette Smoke ............................... 85 Nicotine Absorption ............................................ .85 Alteration of Enzyme Systems ............................. .87 Catwholamine Responses ..................................... .87 Cardiovascular and Belated Effects ....................... .89 Pulmonary Effects ............................................... 90 Fat Metabolism ................................................... 90 Hyperglycemic Effects ......................................... 90 Other Central Nervous System Effects.. ................ .92 Metabolism of Nicotine ......................................... 92 Metabolic Products in Test Animals from Nicotine in Cigarette Smoke.. ........................................ .93 14-4 Related Alkaloids and Their Metabolites in Cigarette Smoke.. ........................................ .94 Pharmacodynamics ............................................... 94 Summary ........................................................... 99 References ........................................................ 100 Reductions of the Toxic Activity of Cigarette Smoke ... 104 Gas Phase ........................................................ 104 Carbon Monoxide ........................................ 104 Reduction of Ciliatoxic Smoke Compounds.. .... 104 Volatile Phenols and Catechols ..................... 106 Volatile N-Nitrosamines ............................... 107 Particulate Phase .............................................. 108 Tar .......................................................... 108 Nicotine .................................................... 108 Polynuclear Aromatic Hydrocarbons ............... 109 Nonvolatile N-Nitrosamines .......................... 112 Polonium-210 .............................................. 113 Summary ......................................................... 113 References ........................................................ 115 Future Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 LIST OF FIGURES Figure l.-Cigarettes: production and tobacco used, 1964 to 1975 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Figure 2.-Tobacco use, 1970 and 1975, men and women, 21 and Over . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Figure 3.-Common tobacco alkaloids and tobacco-specific nitrosamines in cigarette smoke.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 Figure 4.-Tobacco isoprenoids ................................... .50 Figure 5.-Some tumorigenic PAH in tobacco smoke.. ... .53 Figure 6.-Carcinogenic aza-arenes in tobacco smoke . . . . . -55 Figure `I.-Weakly acidic compounds in cigarette smoke.. 56 14-5 Figure 8.--Residues of agricultural chemicals in tobacco and cigarette smoke . . . . . . , . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62 Figure 9.-Degree of protonation of nicotine in relation to pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Figure lO.-Carotid blood levels of nicotine in ng/ml, after the presence in the mouth for 10 minutes of buffered solutions of nicotine at pH 6, pH 7, and pH 8 ..,........ 86 Figure Il.-Mean plasma norepinephrine and epinephrine concentrations in association with smoking and sham smoking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Figure 12.-Arterial blood levels of %-nicotine and W- cotinine, heart rate, and blood pressure during and after smoking a cigarette labeled with `%-nicotine, and during and after intravenous administration of W- nicotine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Figure 13.-Nicotine metabolism.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93 Figure 14.-Structural formulas of some tobacco alkaloids I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Figure 15.-Sales-weighted average "tar" deliveries of U.S. cigarettes from 1957 to the present . . . . . . . . . . . . . . . . . . . . . . . . 109 Figure 16.-Sales-weighted average nicotine deliveries of U.S. cigarettes from 1957 to the present.. . . . . . . . . . . . . . . . . 111 Figure 17.-Benzo(a)pyrene in the smoke condensate of a leading U.S. nonfilter cigarette . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 LIST OF TABLES Table l.-U.S. tobacco production in 1964, 1968, and 1975 by types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Table 2.-Classes and types of tobacco established by the U.S. Department of Agriculture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 14-6 Table $.-Approximate composition of freshly harvested tobacco leaves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Table 4.-Range of chemical composition of tobacco being used in cigarettes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Table 5.-Stalk positions and leaf characteristics.. . . . . . . . . . .18 Table 6.-Stalk positions and smoking properties . . . . . . . . . . . . 18 Table 7.--Representative analyses of cigarette tobaccos.. .21 Table 8.-F&Representative analyses of cigar tobaccos.. . . . . . .22 Table 9.-Correlations among smoke and leaf variables. . . 24 Table lo.-Correlations among selected leaf and biological variables ..,............................................................ 25 Table Il.-Correlations among selected smoke and biological variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Table E-Percent distribution of cigarette smoke . . . . . . . . . 37 Table 13.-Typical mainstream smoke mixture.. . . . . . . . . . . . . .37 Table 14.-Major toxic agents in the gas phase of cigarette smoke (unaged) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Table 15.-Tumorigenic PAH in cigarette smoke.. . . . . . . . . . .54 Table 16.-Major phenols in cigarette smoke ................ .57 Table 17.-Free fatty acids in cigarette smoke ............. .58 Table 18.-Metals in cigarette smoke particulate.. . . . . . . . . . .59 Table lg.-Harmful constituents of cigarette smoke particulate matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Table 20.-Known tumorigenic agents in cigarette smoke particulates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Table 21.-Relative molar potency of nicotine and other cigarette smoke alkaloids.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .96 14-7 Table 22.-Effects of various forms of air dilution on carbon monoxide and carbon dioxide deliveries.. . . . . . . . . 105 Table 23.-Vapor phase constituents with high ciliatoxic potency-in vitro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Table 24.-Removal of some gas-phase components of cigarette smoke by an activated carbon filter.. . . . . . . . . . 10'7 Table 25.~Some measures for `tar` reduction in cigarette smoke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Table 26.-Reduction of biological activity of cigarette smoke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 14-8 Introduction Our understanding of cigarette smoke-its generation, physical composition, toxicity, pharmacology, behavioral effects, and techniques to modify its composition-has advanced considerably since the last review on cigarette smoke in the 1972 report on The Health. Consequences of Smoking. Technology has played an important role in advancing our under- standing of cigarettes and their resulting smoke. One aspect in particular that has improved our understanding is the development of new instrumentation and miniaturization of analytical tools. For example, Baker (I) reported on the use of a fiber-optic probe system for determining and differentiating solid and gas temperatures within the coal of a burning cigarette. The advance made it possible for Osdene (5) to define more clearly the reaction mechanisms that occur in the burning cigarette. Such information should make intelligible modification of cigarettes and cigarette smoke more of a science and less of an art. Another example has been the development and refinement of the Thermal Energy Analyzer, which allows scientists to quantify the level of N-nitrosamines in cigarette smoke (2, 3). The development of reconstituted tobacco sheet technology, designed, at least in part, for better utilization of the tobacco plant in cigarette manufacture, has given manufacturers additional control over the delivery of certain constituents of cigarette smoke, permitting alteration of the combustion process and consequently the levels of smoke condensate produced (4. In this chapter we will consider the tobacco as a raw material, how it is made into.cigarettes, the cigarette smoke generation process, the composition of cigarette smoke, physiological responses to cigarette smoke, the pharmacology of nicotine as a component of cigarette smoke, and efforts to define less hazardous cigarettes through cigarette smoke modification. Also, consideration will be given to the effects of smoke characteristics on smoking behavior and, therefore, on the dose inhaled by man and experimental animals. 14-9 I Introduction: References (I) BAKER, RR. Temperature distribution inside a burning cigarette. Nature 247: 405406, February 8,1974. (2) BRUNNEMANN, K.D., YU, L., HOFFMANN, D. Assessment of carcinogenic volatile N-nitrosamines in tobacco and in mainstream and side&ream smoke from cigarettes. Cancer Research 37(g): 3218-3222, September 1977. (3) FINE, D.H., RUFEH, F., LIEB, D., ROUNBEHLER, D.P. Description of the thermal energy analyzer (TEA) for trace determination of volatile and nonvolatile N-nitroso compounds. Analytical Chemistry 47(7): 1188-1191, June 1975. (4 MATTINA, C.F., SELKE, W.A. Reconstituted tobacco sheeta. In: Wynder, E.L., Hoffmann, D., Gori, G.B. (Editors). Proceedings of the Third World Confer- ence on Smoking and Health, New York, June 2-5,1975. Volume 1. Modifying the Risk for the Smoker. U.S. Department of Health, Education, and Welfare, Public Health Service, National Institutes of Health, National Cancer Institute, DHEW Publication No. (NIH)%-1221,1976, pp. 67-72. (5) OSDENE, T.S. Reaction mechanisms in the burning cigarette. In: Fina, N. J. (Editor). The Recent Chemistry of Natural Products, Including Tobacco: Proceedings of the Second Philip Morris Science Symposium. New York, Philip Morris, Inc., 1976, pp. 4259. 14-10 The Cigarette: Composition and Construction Tobacco, a member of the nightshade family (28), is an important agricultural and economic crop that is produced in almost all parts of the world and used in nearly every country. The tobacco plant Nicotiana tubacum L. is a native plant of the Americas and is used primarily for the manufacture of cigarettes, cigars, pipe tobaccos, and to a lesser extent for oral consumption. Its dominance for smoking use is generally attributed to a few of its combustion products which induce physiological effects to be discussed later in this chapter. The tobacco plant is an excellent material for research in plant and biological science (21). The characteristics of tobacco smoke are primarily functions of the physical and chemical properties of the leaf; hence, one can approxi- mate the levels of nicotine, tar, and other smoke components based on certain physical and chemical properties of the leaf (32). Wide variations in botanical, chemical, and physical characteristics of leaf tobacco are found among the various species, types, varieties, strains, and grades; the quality of the tobacco leaves is predetermined by genetic makeup and subsequently influenced by weather conditions, cultural practices, soil properties, curing, and other post-harvest handling practices (27). The relatively sweet Orinoco-type tobacco, Nicotiuna tubacum L. was successfully introduced for cultivation in Jamestown, Virginia in 1611 and into Europe, Asia, and South Africa by the early part of the 17th century. Worldwide production has increased in recent years (26). During the years 1973 through 1975, worldwide total acreages of tobacco harvested were 10.1,10.5, and 10.7 million acres; yields per acre were 1,054,1,030, and 1,033 pounds; and total production was 10.7, 11.4, and 11.7 billion pounds, respectively (26). Asian countries lead the world in tobacco production followed by North America, Europe, and South America (26). The highest yield per acre appears to be in the People's Republic of China, followed by the United States. The U.S. production for all types of tobacco in 1975 was 2.19 billion pounds. Table 1 summarizes U.S. tobacco production. Since 1964, when the first Surgeon General's Report on Smoking and Health was published, there has been a gradual and continued increase in the number of cigarettes manufactured in the United States (35). It should he noted, however, that per capita consumption has decreased from 11.53 pounds in 1964 to 9.14 pounds in 1975, and total tobacco consumption has declined from 1.41 billion pounds in 1964 to 1.35 billion pounds in 1975. This reduction is due largely to the reduced waste of the tobacco biomass. These results are described in Figure 1. Figure 2 describes the tobacco use for men and women 21 and older for the years 1970 and 1975. It should be noted that there was an 14-11 TABLE l.-U.S. tobacco production in 1964, 1968, and 1975 by types Yield Type and crop year Acreage per Production acre Flue-cured (Types 11-14) 1964 1963 1975 Fire-cured (Types 21-23) 1964 1963 1975 Burley (Type 31) 1964 1963 1975 Maryland (Type 32) 1964 1963 1975 Dark air-cured (Type 3537) 1964 1963 1975 Cigar filler (Type 41-44) 1964 1968 1975 Cigar binder (Type 51-55) 1964 1963 1975 Cigar wrapper (Type 61-62) 1964 1963 1975 Puerto Rican Filler (Type 46) 1964 1963 1975 Total U. S. tobacco (Types ll-72*) 1964 1963 1975 l.cmo acres 628 2,211 533 1,341 717 1.973 32 1,716 23 1,689 23 1,601 307 w= 238 2,372 282 235 31 23 14 14 9 13 14 13 5 31 6 3 1,109 885 1,090 pounds l,l= 1,100 1,050 1,735 1,757 l,@O 1,683 1,766 1,663 1,362 1,321 1,351 l,=O 1343 1.409 1231 lzll 1XQ 2,014 1,941 w@4 million Its. G= 931 1,415 55 39 37 620 563 63 42 32 25 24 19 15 52 41 23 *Includea Perique SOURCE: U.S. Department of Agriculture(S5). increase in the percentage consumption for males and females under 21 years old. Cigarettes are by far the largest single tobacco product. 14-E? CIGAREITES: PROOUCTlON AND TOBACCO USED FIGURE L-In the United States flue-cured tobacco is the most important domestic type, with burley in second place. Note that cigarette production has increased while the tobacco used has remained about the same since 1964. This is due to use of stems, reconstituted sheets and filters in cigarette manufacture in recent years - formerly discarded as "waste". SOURCE: Tao, T.C. (.W TOBACCO USE 1270 AWD 1975 uulmdWommrh21urlOver FIGURE Z.-Use of tobacco by men for cigarettes, cigars, pipes, chewing tobacco and snuff all showed a decrease in the 5-year period 1970-75. Use of tobacco by women also showed a slight drop in cigarettes, but a slight increase in use of cigars and pipea SOURCE: Tao, T.C. (e7). Types and Classes of Tobacco There are at least 65 species within the genus Nicotiuna. The species 14-13 Nicotiam tabacum L. is the main commercially grown species. This species has been established as a natural hybrid between N. Sylvestris and N. Otqvhora (37). The types of tobacco generally used in smoking products are bright (flue-cured), Burley, Maryland, and cigar tobaccos, as well as oriental (aromatic) tobaccos. These types make up the bulk of the tobacco products (Table 1). Other types of tobacco exist, such as Perique, Latakia, and several Indian types, but they are not generally used in U.S. tobacco blends. Over the years, new varieties of bright, Burley, and other tobaccos have been developed that are multipledisease resistant to specific tobacco diseases (23, 28). Within the species of N. &urn, many varieties and types show wide differences in their chemical composition (28). Numerous germ plasms are available in the USDA collection, including approximately 1,000 tobacco introductions, 400 established varieties, and 100 breeding lines. Tso (30) reported that, in a preliminary examination of randomly selected samples from tobacco introductions, there was a threefold variation in sterol content, a tenfold variation in nitrate content, a thirtyfold variation in alkaloid content, and a fivefold variation in phenolic content. He concluded that greater variations probably exist among types not yet studied. Based on methods of curing and the cultivar (a variety of tobacco within a tobacco type) used, leaf tobaccos produced in the United States are separated into the major classes shown in Table 2. There are five classes of air-cured tobacco including light air-cured, dark air- cured, and three kinds of cigar tobaccos: filler, binder, and wrapper (26, 28). Filler is tobacco that makes up the bulk of a cigar, and wrapper is used for the outside covering. Binder is now used primarily for scrap chewing. Binding material for cigars is now made from reconstituted tobacco sheet (RTS). (RTS is also used in the manufacture of cigarettes, as will be discussed later.) Each of these tobaccos has specific characteristics and is produced for a specific purpose. Under class, the subdivision is "types" (26, 2r), based on location of production, method of culture, and in most cases, plant cultivar. The cured leaf from each type is further subdivided into grade groups named on the basis of either principal use in manufacture or stalk position under the U.S. Government grading system. Each of the subdivisions is composed of several grades, determined by several elements of quality, such as body, texture, and color. Physical and Chemical Characteristics In addition to the genetic makeup, environmental factors, including mineral nutrition, soil properties, moisture supply, temperature, and light intensity, affect the chemical composition and physical properties of the leaf (26, 28). The relationships among these factors and the 14-14 TABLE 2-&uwes and types of tobacco established by the U.S. Department of Agriculture Type of curing and class Type no. Type name or locality Flue-cured, Class 1 Fire-cured, Class 2 Air-xmed Class 3A (light air-cured) Class 3E (dark air-cured) Class 4 (cigar filler) Class 5 (cigar binder) Clsss 6 (cigar wrapper) Misdlanaous. Class 7 11A Old Belt-Virginia and North Carolina 1lB Middle Belt-Virginia and North Carolina 12 Eastern North Carolina 13 Border Belt-Southeastern North Carolina 14 21 P and South Carolina Georgia and Florida Virginia Kastern-Kentucky and Tennessee Western-Kentucky and Tennessee 31 Burley 32 Maryland 35 One-Sucker 36 Green River 37 Virginia Sun-Cured 41 Pennsylvania Seedleaf, or Broadleaf 42 Cebhadt 43 Zimmer Spanish 44 Little Dutch 46 Puerto Rico 51 Connecticut Broadleaf 52 Connwticut Havana Seed 53 New York and Pennsylvania Havana Seed 54 Southern Wiinsin 55 Northern Wisconsin 61 Connecticut Valley Shade-Grown 62 Georgia and Florida Shade-Crown 72 Louisiana Perique 77 Domestic Aromatic SOURCE: U.S. Department of Agriculture (36). tricarboxylic acid (TCA) cycle help define the smoking quality of tobacco leaves (3). Smoking quality of tobacco leaf is determined to a great extent by the balance between the carbon and the nitrogen fractions (28). Atmospheric COZ is assimilated by the tobacco leaf through photosyn- thesis, while nitrogen is accumulated by the roots from the soil. The net result of nitrogen assimilation is, therefore, the utilization of a portion of newly photosynthesized carbon chains into the nitrogenous pool. Thus, when the nitrogen supply is abundant, more amino acids and nicotine and less sugar and starch will be synthesized. If the nitrogen supply is limited, acetate will accumulate from the TCA cycle and increase the production of carbohydrates, fats, volatile oils, resins, and polyterpines (26,28). These variations will effect the resulting leaf 14-15 TABLE 3.-Approximate composition of freshly harvested tobacco leaves Constituents Bright cigarette tobacco Cigar filter Carbohydrates Protein Soluble N compounds Inorganic9 Cellulose and lignin Pentosans Pectins Ether-soluble resins Tannins Organic acids Not identified I % 23.0 3.0 122 17.3 3.3 6.7 12.0 14.0 10.0 9.5 20 3.0 7.0 7.0 7.5 7.0 20 25 13.0 13.0 8.0 17.0 SOURCE: Fnnkenburg, W.C. (7). texture, color, porosity, and combustibility. Examples include those tobaccos used in cigarette production, Turkish and bright (flue-cured), as well as cigar tobacco types. The Turkish tobacco is produced with limited supplies of nutrients and water, thus giving leaves more hydrocarbons and highly aromatic qualities (26). Cigar tobacco is grown with an abundant nitrogen supply yielding leaves high in protein and nicotine levels. Flue-cured tobacco is intermediary but slightly toward the carbon side. Table 3 illustrates typical differences among major constituents of bright and cigar tobacco leaves at harvest, and Table 4 describes the ranges of various constituents of the four main tobaccos used in cigarette produetion. Other environmental factors, such as the time of topping and the amount of sunshine (273, also play a role in the carbon-nitrogen balance. The lower right portion of Figure 1 indicates that bright (or flue- cured) tobacco is the most widely used domestic type in the United States, while Burley, a light, air-cured type, ranks second in importance. Together, they account for most of the tobacco used. Typiwl values are flue-cured (45-75 percent), Burley (i545 percent), Turkish (5-13 percent), and Maryland (l-7 percent) tobaccos (26). Some RTS is also used (15-17). The Standard Experimental Blend (SEB) used in the National Cancer Institute's experimental cigarettes, based on 1970 sales-weighted averages; are comparable (25-17). The physical and chemical characteristics of tobacco leaf and smoke are- unavoidably related to one another. Recent studies, particularly with bright tobaccos, show that characteristics such as leaf thickness, rate of leaf burn, and moisture content are significantly correlated with combustibility. Factors that promote good burning will generally 14-16 TABLE I.-Range of chemical composition of tobacco being used in cinarettes* Constituents Flue-cured Burley Maryland Oriental Total nitrogen Protein nitrogen o-Amino nitrogen Nicotine Petroleum ether extmctive Starch Soluble sugars Nonvolatile acids** Water-soluble acids" pH (not %) 1.00-3.00 0.4c1.30 0.08445 O.W.50 3.0&7.50 1.75-3.00 6.0&3200 9.W%.oo 2w-5.06 4.4c5.70 1.50-4.50 0.50-240 O.IO-O.50 0.4CM.50 250-6.00 0.50-3.CKi 0.10-1.50 15.00-36.00 0.3L3.50 5.20-7.50 1.2.5-3.00 0.w1.50 0.084.36 0.65-200 3..5M.50 1.00-3.50 0.50-1.50 13.0@25.00 0.4C3.50 5.3lL7.00 1.4C-3.50 0.7~130 0.10-0.54 0.50-1.30 3.50-7.00 1.90-10.00 3.00-le.00 16.&73.00 4.30-5.25 'Ranges in %. *`Milliliters of 0.1 Nalkali per gram tobacco. SOURCE: Darkis. F.R (S). result in lower levels of TPM in smoke, lower nicotine, cresols, volative phenols, hydrogen cyanide, and benz(a)anthracene, but will yield higher levels of acetaldehyde, acrolein, and carbon monoxide. The position of tobacco leaves on the stalk is known to influence greatly the resultant smoke characteristics (37). Present evidence shows that for higher leaf positions on the stalk, the combustibility is lower, the filling value of the tobacco is less, and the TPM, nicotine, HCN, volatile phenols, and polynuclear aromatic hydrocarbons in the mainstream smoke are higher. Thus, stalk position is an important indicator of both physical and chemical properties of the leaf and aids in interpreting precursors of the final product between leaf and smoke components. Table 5 shows some typical relationships between leaf characteristics and position on the stalk (8, 26, 37'). Table 6 relates the effect of stalk positions and smoking properties (27). Similar data have been described by Wolf (3~). Culture and Harvesting F'ractices Wolf (37) has reviewed the practices employed in tobacco culture and harvesting. A standard field practice with all domestic types of tobacco plants (except shadegrown cigar wrappers) is topping (removal of early blossoms) and suckering (removal of secondary buds) to promote the proper development in leaf size and thickness. Priming (the removal of mature leaves at successive intervals) results in the maximum yield and quality from tobacco plants since leaves at different stalk positions mature at different stages. Depending on the type of tobacco plant and the weather conditions during harvest, there may be as many as nine primings. Stalk-cutting is another method of harvesting, involving cutting the plant at the lowest stalk position and harvesting the entire plant at one 14-17 TABLE 5.-Stalk Do&ions and leaf characteristics Properties of Tobacco Types Lower Leaves Middle Leaves Upper Leaves' Flue-cured tobacco Cell membrane substances Total sugar Total acid o-amino N Nicotine Water-soluble N, total N Soluble ash Tannins, resins PH Air-cured Burley Color Porosity Density Ammonium N, amino N, amido N Nicotine N Comparatively Higher Lower Higher Higher Lower Medium Higher Lower Higher Comparatively Lower Higher Lower Lower Medium Lower Lower Higher Lower Lighter Darker More Less Lighter Heavier Lower Lower Medium Medium Comparatively Lower Lower Medium Higher Higher Higher Medium Higher Lower Darker Lea Heavier Higher Higher *Not including uppxmo& tips. SOURCE: Harlan. W.R. (a), Tso. TX. (27). TABLE O.-Stalk positions and smoking properties Smoking properties Lower leaves Upper and middle leaves Strength (N compounds) Aromaticity (tannins, resins) Mildness (sugars, starch, oxalic acid) and sharpness (cell membrane substances, ash constituents. citric acid) relatively light aromatic somewhat sharp relatively strong highly aromatic mild SOURCE: Harlan, W.R(B),Tso.T.C. (27). time. In general, Burley and Maryland tobaccos are harvested by stalk- cutting. The application of herbicides to control weeds, fertilizers to enhance plant growth, pesticides to treat soil and control plant diseases, and insecticides may directly or indirectly leave residues on plant material; this factor must be considered when the characteristics of the tobacco leaf and smoke chemistry are examined. Curing and Aging The green tobacco leaf primed from the plant goes through a process known as "curing" in order to develop desirable taste and aroma for 14-18 smoke products. Several different curing processes are used to produce leaf tobacco suitable for the manufacture of a variety of tobacco products (37). Curing is a process during which chemical conversions take place in the tobacco leaf. During flue-curing or air-curing, chemical conversion is dominated by hydrolytic enzymes. Disaccharides and polysaccharides are hydrolyzed to simple sugars; proteins are hydrolyzed to amino acids which undergo subsequent oxidative deamination; pectins and pento- sans are at least partially hydrolyzed to pectic acid, uranic acid, and methanol. A second step occurs only in air-cured tobaccos and includes conversions such as the oxidation of simple sugars to acids, the oxidation and polymerization of certain phenolic compounds, and some decrease in alkaloids and dry weight (26). As a result of years of research, numerous advances have been made in the procedures used to harvest, cure, and process tobacco. One particular development in the early 1950's was the process of manufacturing reconstituted tobacco sheets (out of tobacco scrap) in a manner analogous to paper manufacture (13). The process will be discussed later. The significance of the process lies in the fact that tobacco need not be harvested and cured in whole leaf form, thus suggesting new mechanized approaches to harvesting and curing. A new curing procedure called homogenized leaf curing (HLC), developed by scientists at the U.S. Department of Agriculture, involves the homogenization, incubation, and dehydration of tobacco leaf (.4,3X'). The fundamental concept is to cause the necessary chemical changes to occur in a homogenized tobacco slurry instead of in the harvested whole leaf. The process saves considerable hand labor normally required for handling whole leaf, allows a mechanism for removal of undesirable components, and permits better control and enhancement of biochemical and chemical changes. Results have shown that the HLC method may provide smoking quality that is comparable to conventionally cured leaf but with a relatively lower biological response (33). Cured, unaged tobacco is still unsuitable for manufacturing into tobacco products because it has a sharp, disagreeable odor and an undesirable aroma and produces irritating smoke with unacceptably harsh flavor (26). To improve these conditions, cigarette tobaccos (flue- ~ufed, Burley, Maryland and Turkish) are subjected to a further process called aging. Aging greatly improves the aroma and other qualities desirable in smoking products. The aging process can be natural or forced, depending upon time, temperature, and humidity. A l- to Z-year aging period is notunusual for cigarette tobaccos. The treatment of cigar tobaccos consists of two steps (7). The first step is storage and the second is fermentation. Current knowledge of the chemical conversions during aging and fermentation is rather limited (26). The most noticeable chemical changes in the aging process 14-19 are an increase in volatile acids and a decrease in a-amino nitrogen. Flue-cured and Turkish tobaccos also exhibit a loss of reducing sugars and volatile bases other than nicotine. In fermentation, new chemical reactions appear and ongoing reactions are intensified. A decrease in tobacco alkaloids, especially nicotine, is evident (7). Large amounts of ammonia are produced, and amide and a-amino nitrogen levels are decreased. The pH increases because of the elimination of organic acids through oxidation and decarboxylation. It is likely that enzymes, microorganisms, and catalysts all play a part in the fermentation process (26). Representative analyses of aged and cured cigarette and cigar tobaccos are shown in Tables `7 and 8. These chemical variations are.the results of different varieties, cultures, fertilizers, soils, climates, and post-harvesting practices as described above. Other Factors Leaves from different levels on the stalk possess considerably different chemical and physical properties. For example, upper leaves possess higher nicotine, lower total sugar, higher tannins and resins, lower ash, and higher total nitrogen; lower leaves tend to contain higher total acid, higher soluble ash, and higher pH. However, not all substances are at their highest or lowest concentration in the upper and lower leaves. The leaves at the middle stalk position, for example, have the highest sugar, lowest a-amino nitrogen, lowest total acid, lowest total nitrogen, and lowest soluble ash. Selecting mature leaves at various time intervals (priming) allows maximum use of tobacco leaves and selectivity in future blending. Because of the chemical and physical differences, leaves from various stalk positions also vary in smoke characteristics, as shown in Tables 5 and 6. Lower leaves usually deliver a lighter "strength," somewhat sharper taste, and less aromatic smoke than the upper and middle leaves (1). These smoking properties are largely functions of chemical composition. For example, nitrogen compounds are believed to be associated with strength; tannins and resins are associated with aromaticity; sugars, starch, and oxalic acid are associated with mildness; and cell membrane substances, ash constituents, and citric acid are associated with "sharpness" (I). Certain physical quality factors are also related to chemical components, as all these variables are interrelated. In a recent study with bright tobaccos (31), many physical variables including leaf thickness, rate of burning, leaf color, moisture content, moisture equilibrium, specific volume, and t&home numbers were found to be significantly correlated with many leaf chemical variables. The presence of radioelements, including radium-226, lead-210 and polonium-210 have been reported in tobacco and tobacco smoke (19) and reviewed recently by Harley and coworkers (9). Contents of Po210in 14-20 TABLE 7.-Representative analyses of cigarette tobaccos (leaf web after aging, moisture-free basis) Component ?& Flue-Cured. Burley. Type 13 Type 31 Maryland. Type% Turkishb Total volatile bases as ammonia Nicotine Ammonia Glutamine a3 ammonia Asparagine as ammonia a-Amino nitrqen as ammonia Protein nitrogen as ammonia Nitrate nitrogen as NOs Total nitrogen as ammonia PH Total volatile acids as acetic acid Formic acid Malie acid Citric acid Oxalic acid Volatile oils Alcohol-soluble resins Reducing sugars as dextrose Pectin as calcium pectate Crude fiber Ash calcium as CaO pota9sium as K2.0 magnesium as MgQ chlorine as Cl phosphonrs as P& sulfur as SOI Alkalinity of water-soluble ash C 0.282 0.621 0.366 0.289 1.93 2.91 1.27 1.05 0.019 0.159 0.130 0.105 0.033 0.035 0.041 0.020 0.025 0.111 0.016 0.058 0.065 0.203 0.075 0.118 0.91 1.77 1.61 1.19 trace 1.70 0.087 trace 1.97 3.96 2.80 2.65 5.45 5.80 6.60 4.96 0.153 0.103 0.090 0.194 0.059 0.027 O.CB 0.079 2.83 6.75 243 3.87 0.78 8.22 298 1.03 0.61 3.04 2.79 3.16 0.148 0.141 0.140 0.248 9.08 9.27 8.94 11.28 22.09 0.21 0.21 12.39 6.19 9.91 12.41 6.77 7.88 9.29 21.79 6.63 10.81 24.53 21.98 14.78 2.22 8.01 4.79 4.z 2.47 5.22 4.40 2.33 0.36 1.29 1.03 0.69 0.84 0.71 0.26 0.69 0.51 0.57 0.53 0.47 1.23 1.98 3.34 1.46 15.9 36.2 36.9 s.5 `In % except for pH and alkalinity. "Blend of MPEedonia, Smyma, and Samsun types. +fillilitem of IN acid per 100 g tobacco. SOURCE: Harlaa. W.R (8). leaf tobacco and tobacco soil vary with the origin of the sample and methods of culture and curing (24). Polonium seems not to be entirely derived from radium. The plant probably takes it up from the soil or air. The general range of PO210 in tobacco leaf varies from 0.15 to 0.48 pCi/g (10-U Curies per gram); in tobacco-growing soil, it varies from 0.26 to 0.55 pCi/g. The amount of Ra-226 in tobacco-producing soil appears to be related to phosphorus fertilization. Soils having high available P continuously used for tobacco crops usually have a higher FL226 content, the range being 0.52 to 1.53 pCi/g (24). The significance of these radioelements in tobacco and tobacco smoke is being extensively studied with P&lo-enriched leaf tobacco by USDA. 14-21 TABLE &--Representative analyses of cigar tobaccos (leaf web after fermentation, moisture-free basis) Corm. shade- Northern Puerto n Lo n Wisconsin Penn Brown binder. filler. -pper. Type& Type 41 Type 61 Total volatile bnsea as ammonia Nicotine Ammonia Total amide BS ammonia Pmtein nitrogen as ammonia Total nitrogen as ammonia PH Ash Alkalinity of water-soluble ashb 1.293 1.055 0.874 0.707 1.47 268 204 0.90 0.914 0.575 0.465 0.348 0.2% 0.199 0.165 0264 220 214 288 3.26 5.18 4.75 5.16 4.65 5.33 5.17 627 6.33 6.10 1.31 6.56 7.25 23.79 24.94 34.50 2245 2257 2234 30.4 45.5 47.0 627 43.0 33.6 1.478 2.a 1.012 022 281 3.01 0.670 1.43 0.313 0.208 *In 46 except for pH and alkalinity. Vdilliliters of IN acid per 100 g tobacco. SOURCE: Harlan. W.R (8). Aflatoxin BI, the most toxic of the four known aflatoxins, is produced by Aspergillus flavu.~ Lk. ex Fr. The binding of aflatoxin BI to both native and denatured deoxyribose nucleic acid (DNA) partially explains its extreme toxicity and carcinogenicity. Aflatoxins have been reported to occur in many commodities, but its presence in leaf tobacco haa not been positively confirmed, although A. flavus was known to be present in various grades of air-cured Burley tobacco. Certain types of tobacco contain higher populations of fungi than other types (6). These differences probably result from culture, curing, and handling practices as well as from the chemical composition of tobacco leaf and the climate in which it is grown. An examination of samples of leaf tobacco and of cigarette smoke condensate by Tso, et al. (26) failed to show aflatoxin Bl. Pure aflatoxin Bl added to cigarettes was not recovered in the smoke condensate, indicating that aflatoxin BI, even if present, was changed or decomposed during the smoking process. Relationships Among Tobacco Leaf, Smoke, and Biological Response Recent reports have been published dealing with precursor-product relationships among specific leaf tobacco components and smoke constituents (20,26,31,34). One comprehensive study was conducted to examine the relationships among leaf, smoke, and biological responses using well-defined bright tobacco samples specially produced for this 14-22 purpose. This study involved a total of 151 variables, including 102 leaf and agronomic characteristics, 42 cigarette and smoke components, and 7 biological responses (31). The results clearly indicated that certain leaf characteristics could be used as "markers" to predict total smoke delivery or individual smoke components. These findings demonstrated that modification of these markers through genetic, cultural, or curing procedures might lead to the development of leaf tobacco of more desirable quality and usability. The correlations made by Tso and coworkers may be interpreted in the sense of precursor-prooust relationships between specific leaf and smoke components and between certain smoke components and biological responses. Table 9 gives the correlations among some selected leaf and smoke variables. Using the same selected leaf characteristics, the correlations with the results of seven short-term bioassay systems were determined as shown in Table 10. The sebaceous gland suppression system showed many significant and interesting correlations with certain leaf characteristics (34). In examining all these variables, the authors commented that one significant factor appeared to be the one which affects leaf combustibility and thus the formation of components that affect suppression. Variables that promoted combustion were general- ly negatively associated with suppression, and variables that inhibited combustion were generally positively associated with suppression. In addition, phenolic compounds were positively associated with suppres- sion. These compounds may serve as precursors of smoke constituents with tumor-promoting activity. In addition to the sebaceous gland suppression system, the E. coEi., virus-infected quail, and mixed cell-culture systems also used cigarette smoke condensate. These three systems did not demonstrate any meaningful correlations with the variables examined. Correlations among selected smoke and biological variables are shown in Table 11. For example, static burning rate was negatively associated, whereas total phenols, benzo(a)pyrene (BaP), benz(a)anthracene (BaA), and smoke pH were positively associated with sebaceous gland suppression. Tso, et al. (34) commented that it is somewhat surprising that dry total particulate matter, cresols, acetaldehyde, acrolein, and hydrogen cyanide did not show any statistically significant correlation with the biological data employing whole smoke in these studies. Smoke delivery and smoke composition thus seem to depend on the characteristics of leaf tobacco (26). The effects of genetic and stalk position differences are reflected in botanical, physical, and chemical properties of leaf tobacco, which in turn are clearly illustrated in the smoke constituents of these experimental samples. These results agree with those of parallel studies using leaf "markers" for identification of leaf quality and usability as described by Tso and Gori (32). Usability in their definition represents the state of being usable without adverse 14-23 TABLE S.-Correlations among smoke and leaf variablea Acmlein SaP amoked) nmokd) Trichoim Lplf thiiknras Firehddiw up&y Moiture equilibrium pH (leaf totauo) K cell-w.3 ."manee Total N Nitrate N TOW alkaloid (dw.1.l Tot.1 vol. hsea ~1 mine N Total free amino act& Aginine AlpattiC acid Pmlim Dimtthylamme Toti, polyphenols Chlomgenh rid Rulin Smpoktin Limdn oiic cid Malie acid Penladeenoic acid Stigmlsteml p,p'-TDEE Total DDT + TDE Amma FIWW StrPn@h 601.. ,450" -.4W .5sP* ,681" -.6W all" 469" ,680" -.5a6" 615" .154** .X93* -212 -.663** .xw * .367* -.a3 -.526" .S4- -.5x3" ,985" .BoD" ,415,' 445'. 263 - 410. .a3 .x69- -356. -.wY ,364. 459" ,573-e - 474" ,151 .5&.. ,561.' 444' ,141 -.lm- .rn" -.140 .37S* .W" ,516'. .m- -.431** -.uB' ,410' SW -.5fP~ -.346* .xl*~ m .378* -364 .531** -.Tzl .410" .416' 621.' 476.. ..lz? ..5Tl'. ,407. .mB .3sz* .x4** a36 -.306 ,167 46s -359. ai3 .a33 m3 324 ..193 -.llS .161 -.m4 245 -.456- ,016 -.m** .112 m5 ,161'. -.a05 .x4 211 ,313 .x4 Ma- -.153 ,546" .BOB" .lU -4s' 282. -.m 433 -.175 -.5w* -.890** ,459" -.5W- -.I63 -.a36 -.l35'. 466 -.lm)" .4lP -.M Bnr* -331 -.om .?a6 212 on .m5- .54S- -.I65 * .APa** ..634- -.m6- 218 la,`. -A31 .Ml- .lzP* ,566. .505** 687" .Aw* 355. .A63- ,399. .A6%" aa3 ,736" 5m** -.lsB- -457" 567" -.6W' .39- ,519-e ,928 .zal .546- ,144" .?m" ..m- ,659-a -.6W' -.mi** ..433' .918- -.&W .%32- .BM- .496" .mP ,447. ,463' ,126' Ml.' ,493'. ,463" .4&?- ,801.. 99(1 ,546" -.l!a" .YlO" -.soB" .68- ,435.' .SW* .5cw .BBB" TABLE IO.-Correlations among selexhd leaf and biological variables Variable sebaceous E. cdi Virus- Mixed Cilia gland r.one infected cyte Mm inhibition quail cuT:m toxicity toxicity phage Stalk position.. ...................... 0X16'* Ttichome ............................. 391' Leaf thickness ....................... 352' Rate of burn.. ...................... -X4** Moisture equilibrium ................ .466** pH (leaf tobacco) ................... -.494** Potassium.. .......................... -.523* o Total nitrogen ....................... .595** Nitrate nitrogen .................... -.473** Total alkaloids.. ..................... ,439' Total volatile basea ................. 458" a-Amino nitrogen ................... ,178 Total free amino acids.. ........... 255' Aspartic acid ........................ -337 Dimethylamine ...................... .451** Total polyphenols ................... XC? Chlorogenic acid.. ................... 509" Sqoletin ............................ .486** Oxalic acid ........................... ,397' Malic acid.. .......................... -507.' Pentadecenoic acid. ................. ,196 stigmasterol ......................... -.361* Total DDT + TDE.. ............... &Xl** Flavor ................................ .3w Strength ............................. .426* a.030 -0.009 -.169 .007 .06a .156 ,011 -al3 -.lOo ,056 ,104 -264 -. 106 -221 -SE36 200 ,015 ,146 -.W ,219 -LO31 .zB -.%I3 a4 -239 -.012 -.048 -.107 394' -.042 -.223 .143 -.025 ,160 -076 a.4 4339 Ml' -.117 -.072 -.123 ,143 -.070 -.171 .030 .160 -.l26 -.OlO ,147 .048 -0.316 m37 -0.076 0.023 -32-l -.153 -.lll 43s -.313 ,295 -.373' -.004 ,193 -.034 ,017 091 -.4tio** .I43 .oso -.054 209 439 ,154 -.152 ,070 466 -.016 .043 -I94 .037 496 .171 2% .035 .w3 .m -.124 255 -.150 ,166 -.ot!J .140 -.130 .175 .064 -306 400 "247 -.ofJ7 -304 -.lll .63 .172 -.X8 402 ,134 330 ,017 -.133 ,136 -.353* -.197 ..I01 446 -326 ,086 -050 .098 -264 .on -.181 .os5 .02_ -.lrn -.014 .I04 .a4 .zz3 .020 .105 .064 -.375' 274 -.106 -.lOl -.171 225 443 -.166 -271 .lOZ ,159 -.!?A9 465 ml -.178 -272 -la ,144 I26 ' and o * - signifiicrntly different from 0 at 5 and 1 pemnt, mqmtively. SOURCE: Tao, T.C. (2.5). Usability index = A B If chemical, physical and botanical characteristics are considered: A + C+D Usability index = - - B E - nitrate + K + total ash + cellulose, B^ = nicotine + TVB + a-amino nitrogen + starch + polyphenols + PEE + lipid residues + waxa + phytoaterois + fatty acids, C - filling value + combustibility, D - stem/lamina ratio, E = thickness (WB = total volatile bases, PEE = petroleum ether extracts and K - potassium) 14-25 TABLE ll.-Correlations among selected smoke and biological variables Variable' Sebaceous E' CXi $tz& ML; Cilia- Cytc- Macro- gland "Pye. mhlbltlon quail cu,ture toxicity toxicity @age Static burning rate per minute.. ........................ mg-O.465.' Dry total particulate matter'. ....................... g 272 Nicotine in smoke* ........... mg ,268 o-, IR-, and p-Cresols~ ........ mg ,137 Total volatile phenols' ....... mg .542** Acetaldehydel ................. mg -.104 Acrolein' ....................... mg ,973 Hydrogen cyanide'. ........... mg ,138 Benr.+lpyrene' ............... pg .3&J* Henzo[a]anthracene~ .......... Irg ,446 o Smoke pH (last puff) ........ pH .468** Carbon monoxide' ............ mg 285 Carbon dioxide* ............... mg 323 0.010 234 ,073 ,171 204 ,116 -.074 -.165 .054 -.ll2 -.329 -.109 489 ,152 280 ,249 .2Q5 -.@I8 ,291 434 .213 IO5 ,373. ,136 312 -0.145 0.390* .I04 -.013 .a35 -.322 433 ,109 ,163 ,019 -.024 -IO3 .w2 ,031 -0.128 ,272 .472** 243 .Oll -.216 -333 .l25 ,251 -.170 34.5 -444 -So4 -.196 -314 a30 -.018 .145 -.130 .067 .025 zs 428 -.176 1.*4** B aigniicantly different from 0 at 5 and 1 pavent, respectively. `per pm tobacco burned `per 100 grama tobacco bumed SOURCE: Tm. T.C. (OS). effects. Markers were used to establish a "usability index." High emphasis was placed on the chemical constituents, Physical factors were next in importance because they can be improved through reconstitution. Botanical factors were considered only when natural leaf was used and entire stems were returned for cigarette manufac- ture. Thus, the potential is there to assume that modification of the markers identified in this type of analysis may lead to the improve- `ment of the smoke products as well as the biological effects of the smoke. Modification of Tobacco and Tobacco Products It has been reported by Tso and coworkers (33) that the labor of tobacco harvest and post-harvest handling may account for 50 to 55 percent of the total required to produce the crop. Consequently, many attempts have been made to reduce use of hand labor. It is not essential that the tobacco leaf be kept whole in order to be useful to the tobacco industry (14). Tso and coworkers (4, 33) recently reported the results of a new procedure for curing leaf tobacco through homogenization, incubation, and dehydration, called homogenized leaf curing (HLC). The objectives of the HLC process were threefold: to reduce production labor costs, to reduce or eliminate undesirable factors that may be associated with the smoking and health problem, 14-26 and to improve tobacco usability by enhancing certain physical and chemical factors. Preliminary results (4, 33) suggest HLC advantages are the capability for more complete mechanization and the enhanced potential for reduction or elimination of substances found to be hazardous to health. Reductions in total volatile bases, nicotine, reducing substances, total particulate matter, and nitrosamines have been reported (33). Another method of modifying tobacco and tobacco products involves development of the reconstituted tobacco sheet (RTS); this method has been reviewed by Moshey (14) and Mattina and Selke (13). The original impetus for developing a reconstitution process was purely economical. For each pound of auction weight tobacco, only about 63 percent was usable shredded leaf tobacco, although approximately 6 percent of the stem material was also blended in smoking tobacco. The remaining 31 percent, consisting of sand (2 percent), discarded stems (18 percent), manufacturing fines (1 percent), and moisture and aging loss (10 percent) was lost to the manufacturer. A process that could utilize the lost stems and fines and control moisture would increase the amount of usable tobacco from a harvest, cut costs, and offer some manufactur- ing control over the physical and chemical properties of the resultant product (13). Several processes were developed in the early 1950's. These were of two general type groups; in one group, the tobacco is ground into fine particles, mixed with a hydrocolloid gum, and cast on an endless steel belt. The other, more widely used group of processes, involves mechanically working the insoluble portion of the tobacco into a fibrous mass and forming it, via paper-making techniques, into a web. In one variation of the paper process, the soluble portion is diverted prior to the paper-making and then added back to the self-supported web. In another variation, the soluble portion remains with the fibrous material throughout the processing. For all processes, the finished product is in the form of leaflets which are then blended with natural tobacco and shredded. The significance of the sheet process lies in the ability to chemically and mechanically produce desired changes during the pulping process. For example, chemical extractions can be performed to reduce nicotine and other constituents. Tar-yield levels can be reduced to some extent, and additives can be put into the material. The structural modifica- tions which can be effected through reconstituted sheet technology could result in considerable differences in the burn properties and in the smoke. Produced tobacco sheet with a 10 mg/cigarette tar yield without filtration is now available using RTS technology. Lower figures are possible but may cause the sheet to be undesirable as a tobacco product. Flavorings and other additives can also be added at selective stages during the process if necessary, depending upon the solubility and volatility of the additive. The components of leaf tobacco can be classified into three different categories.- Some components are essential for smoke quality and desirability, others have either little or no effect, and a third category consists of components that serve as precursors of undesirable smoke constituents such as HCN and aza-arenes (5,28). One class of components in the third category is fraction-l-protein (12,28,29). This and other proteins do not contribute in any significant way to smoke aroma or flavor. Removal of fraction-l-protein achieves two purposes-improved leaf quality and usability, and fraction-l- protein as a potential food source. It is estimated that up to 6 percent of the tobacco yield could be used for feed and food purposes (28). Fraction-l-protein is the major soluble protein of green plants and may account for 50 percent of the soluble protein fraction and 25 percent of the total protein (26, 28). The protein is an enzyme called carboxydismutase (21) that catalyzes the first step in the transforma- tion of CO2 into carbohydrates during photosynthesis (28). Tso (33) and DeJong (4) have reported that the fraction-l-protein can be removed for beneficial use by the above-mentioned HLC process, and could be used as a food source for millions of people annually (28). The protein has been evaluated as a food source (28, 29) and found to compare favorably with egg and human milk for essential amino acid content. Cigarette Engineering The tobacco blend can vary in the amount of Burley, bright (Virginia), Maryland, and oriental leaf and in the amount of reconstituted tobacco sheet used. Casing solutions are used to hold the tobacco blend together. Humectants (moisture retainers) are added to maintain the necessary body and moisture qualities and to contribute to the flavoring of the blend. Flavor-enhancing additives are used to make the smoke pleasant and more acceptable to the smoker. To maintain the physical integrity of the product, a paper wrapper is used. Each c,f these ingredients may affect the burn rate, puff number, pyrolysis products, and ultimately the chemical constituents of mainstream and sidestream smoke and smoke condensate. Typical casing materials that :ilay be u: ,+I are sugars, sirups, licorice and balsams. These additives imProve or change the flavor characteris- tics and burning qualities and impart important binding qualities to the blend. However, additives, when pyrolyzed, may yield undesirable as well as desirable products. Licorice, for instance, could be a precursor of polyaromatic hydrocarbons (PAH). Sugars used in casings cause an increase in furfural, nicotine, and tar in resulting smoke and a decrease in volatile acids (21). Flavoring agents are added at different steps in the cigarette manufacturing process, depending upon volatility. Volatile flavors. such as alcohol-soluble fruit extractives, menthol oils, and arc?a! IA-9"