Biochemical studies of isolated hamster tracheal epithelium.

The epithelial lining of respiratory air passageways is a primary target tissue for toxicity and carcinogenesis in man and in animal models of human disease. The importance of this target tissue was the basis for development of methods to study its biochemistry, and with this information to distinguish the unique properties of this tissue from properties common to all cell types. Biochemical methods employed labeling of macromolecules in isolated hamster treacheas during brief (less than 4 hr) incubation in vitro. Studies of RNA metabolism in isolated tracheas demonstrated a pattern of maturation of ribosomal RNA like that shown for other cell types. Alterations in RNA metabolism were observed in isolated tracheas obtained from vitamin A-deficient hamsters and hamsters previously treated by intratracheal administration of benzo[a]pyrene (BP) plus ferric oxide (Fe2O3) in vivo. Studies with toyocamycin, actinomycin D, and alpha-amanitin, all inhibitors of RNA metabolism, were performed to characterize the class of RNA molecules with a decreased proportion of labeling in tracheas from vitamin A deficient hamsters. In another series of experiments, BP was shown to bind to DNA in epithelial cells of isolated tracheas. The quantity of BP binding was increased by prior intratracheal treatment of hamsters with BP plus Fe2O3 in vivo, this induced binding was inhibited by addition of 7,8-benzoflavone to the incubation medium. Increased BP binding was also observed in isolated tracheas from hamsters believed to be in states of increased susceptibility to respiratory carcinogenesis in vivo. The results show that biochemical studies are feasible with this tissue. Furthermore, a number of questions of importance with regard to this target epithelium are best studied directly in its constituent cells.


Introduction
That the respiratory tract is a primary site of interaction with the environment is a fact well known to many city-dwelling scientists and laymen alike. Environmental pollution is a fact of life that greets you at your front door as you open it to leave each day. But rather than being another one of the petty annoyances of modern life, respiratory exposure to carcinogens and other environmental contaminants, along with respiratory exposures faced in the home and at the workplace, may threaten our health and is responsible in part for the increasing toll from lung diseases including lung cancer. *Department of Pathology, University of North Carolina Medical School, Chapel Hill, North Carolina 27514.
Rather than the whole body or the whole lung, the critical site of biological interaction with many of these environmental agents is the small mass of tissue constituting the respiratory passageways, and in particular, their epithelial lining cells. This is particularly evident in the case of lung cancer, since the majority of these tumors are presumed to arise from cells of this epithelium. This fact was the basis for the development of biochemical methods for study of a respiratory airway epithelium. The Syrian golden hamster was chosen as the tissue because a method for producing respiratory tract tumors resembling those of man had been developed with this species (1) and because of the near absence of respiratory infections in these animals. The tracheal epithelium was selected since it is most readily obtained and since is is a site at which a high in-cidence of experimental tumors develops (1,2). This report describes studies related to nucleic acids of the tracheal epithelium from normal hamsters, hamsters treated with benzo[a]pyrene and hamsters made vitamin A-deficient. The aim is to show that a number of sophisticated biochemical studies are feasible with this important target tissue. These investigations are an example, rather than the limit, of biochemical studies of this tissue, since other studies have considered glycoprotein synthesis (3)(4)(5)(6)(7)(8), carcinogen metabolism (9), and isoenzyme distributions (10) in the tracheobronchial epithelium.

Materials and Methods
Male random-bred Syrian golden hamsters (Mammalian Genetics and Animal Production, Division of Cancer Therapy, National Cancer Institute) were maintained in pairs on San-I-Cell (Paxton Processing Co.) and given water ad libitum. They were fed either Wayne Lab-Blox (Allied Mills, Inc.) or a vitamin A-deficient diet (General Biochemicals). Vitamin A-deficient hamsters and pair-fed controls were prepared as described previously (11). Benzo[a]pyrene (Aldrich Chemical Co.) was hand ground with an equal mixture of Fe2O3 dust (Type 3098; Charles Pfizer, Inc.) and a suspension of this mixture was instilled intratracheally into some hamsters by methods previously described (1). Hamsters  For preparation of whole cell RNA, tracheal epithelial scrapings from two tracheas were pooled, 100 ,ul of 0.80/ sodium dodecyl sulfate (SDS) was added, and the mixture homogenized at 0°C. Liquified anhydrous phenol (600 jA, 45-50°C) was added, and the mixture was further homogenized for 2 min at 0°C. The resultant emulsion was centrifuged at 10,OOOg to obtain phase separation, and the aqueous phase was saved. This aqueous phase was further concentrated with a second phenol treatment as described above (approximately 30 jA final volume).
Samples containing approximately 5 Mg RNA were fractionated by electrophoresis on composite agarose (0.50/0 )-acrylamide (1.7%) gel slabs as previously described (12). Hamster liver cytoplasmic RNA labeled in vivo for 15 hr with uridine-2-14C was added to each experimental sample to provide internal markers to identify the 18S and 28S species. After electrophoresis, the gels were sliced into 1-mm pieces and the slices placed into scintillation counting vials with 0.1 ml water and 1.0 ml NCS solubilizer (Amersham-Searle). After 2 hr, 10 ml of tolune-PPO-POPOP scintillation counting solution was added to the vials; vials were allowed to stand overnight before counting. Counts per minute were corrected for spillover between channels, efficiency of counting, and background counts to give disintegrations per minute (DPM). The sum of DPM recovered from all slices of a gel was determined and the percentage of this total present in each slice was calculated. These percentages were plotted as a function of the gel slice number, which is equal to the migration distance in millimeters.
For purification of DNA, the epithelial scrapings were transferred to a small Dounce homgenizer, and the suspension was adjusted to 5mM Na3EDTA and 2% SDS. The mixture was homogenized at 0°C and then treated at room temperature with an equal volume of water-saturated phenol. The aqueous phase was separated and saved. A second phenol treatment of the aqueous phase was performed, two volumes of ethanol were added to the final aqueous solution, and the mixture was stored at -20°C overnight. The resulting precipitate was sedimented by centrifugation at 10,000g for 2 min, and the supernatant was discarded. The precipitate was extracted with multiple aliquots of ether until there was no radioactivity present in the extracts, then resid-ual solvent was evaporated in a stream of nitrogen. The precipitate was dissolved by homogenization in 0.1M Tris-HCl, pH 7.5, containing 0.15M NaCl in a Dounce homogenizer. The aqueous solution was then sequentially digested with pancreatic RNase (Worthington Biochemical Co.; final concentration 100 ,ug/ml, previously incubated at 900C for 1 hr to inactivate DNase) at 37°C for 1 hr, and pronase (Calbiochem; final concentration 100 pg/ml) at 37°C for 1 hr. After enzymatic digestion, the samples were dialyzed at 40C during a 3-day interval against multiple changes of 0.1M Tris-HCl, pH 7.5, containing 0.15M NaCl. After dialysis, the sample solutions were adjusted to 5M CsCl and 4.2mM Na3EDTA and the samples were centrifuged at 35,000 rpm for 66 hr at 20°C in a Beckman SW 56 rotor. After centrifugation the CsCl gradients were fractionated into 0.2 ml portions while the absorbance of the effluent stream was continuously monitored at 254 nm. Radioactivity was determined for each fraction and the absorbance of peak fractions was determined at 260 and 280 nm.

Results
Although the majority of our efforts have concerned nucleic acids of tracheal epithelium, our initial studies were designed to document the preservation of morphological and biochemical characteristics of the isolated tracheas. At the limits of our typical in vitro incubation, both histological and ultrastructural features of the tracheal explants were essentially normal. Linear incorporation of leucine into proteins during the period of in vitro maintenance indicated continuity of biochemical function. These results were reported previously (12,13). Autoradiographic studies demonstrated that grains from incorporated leucine-3H were predominantly over mucus cells. Similar results have also been seen in long-term tracheal organ cultures from rats (M. J. Mass and B. P. Lane, personal communication). Intense labeling of cells with incorporated thymidine-3H was infrequent in normal tracheas, but basal cells and nonciliated, nonbasal cells (sublumenal plus lumenal) were nearly equally represented among those cells labeled. Mechanical injury of the trachea in vivo (with a cannulation needle) prior to tracheal isolation increased the numbers of cells incorporating thymidine. Autoradiographic studies of tracheas incubated continuously with medium containing uridine-3H initially (e.g., at 30 min) had grains predominantly over cell nuclei. With longer periods of incuba-tions the distribution of grains between nucleus and cytoplasm become uniform (e.g., 3 hr). Accompanying studies demonstrated that precursors were also incorporated into the supportive structures of the trachea. Since supportive structures were discarded in studies of the epithelium, this incorporation did not influence the observed results.

Studies of RNA Metabolism in Isolated Hamster Tracheas
The first biochemical measure of RNA synthesis which we studied in the tracheal epithelium was the extent of incorporation of uridine-3H into RNA as a function of time. After various periods of incubation in vitro, tracheal scrapings were homogenized and washed with 0.2M perchloric acid (PCA) to remove unincorporated radioactivity (12). The DNA and RNA in washed homogenates were then hydrolyzed with 0.6M PCA at 7000, and radioactivity and DNA content (14) were quantitated. Uridine incorporation (normalized to DNA content) was approximately linear for up to 4 hr ( Fig. 1). Evidence supporting the identity of the radioactivity as RNA were its lability to hydrolysis with 0.3M NaOH (3700 for 60 min) and the marked inhibition (91%) of incorporation caused by preincubation of tracheas in medium containing actinomycin D (5 pg/ml, 30 min). Inhibition of RNA synthesis has been seen in cultured rat tracheal rings incubated with a comparable concentration of actinomycin D (15).
Whole cell RNA was extracted from tracheal scrapings immediately after a 30 min pulse exposure to medium containing uridine-3H and at 30, 90, and 150 min after transfer of tracheas to unlabeled medium. The RNA was separated electrophoretically on acrylamide-agarose gels, and the distribution of radioactivity in species larger than approximately 7S was determined for each of the observation times ( Fig. 2). Uridine-"4C-labeled cytoplasmic RNA from hamster liver was mixed with each sample to provide an internal reference for identification of 28S and 18S species. At 30 min, label was principally in heterogenous high molecular weight RNA but with a distinct peak (at 27 mm migration) consistent in size with the 45S ribosomal precursor. By 60 min, small 28S and 18S peaks were present at the location of the 11Cmarker RNA. In addition to the persistent peak at 27 mm, a new peak was evident at 41 mm consistent in size with the 32S ribosomal precursor. By 120 and 180 min, sizeable 28S and 18S peaks were present. Estimates of the molecular weights of the presumed 45S and 32S Isolated tracheas were incubated in vitro for the indicated time intervals in medium containing 5,uCi/ml of uridine-3H.
Following incubation, the epithelium was scraped from supporting structures of the trachea and unincorporated radioactivity was removed by extraction with cold 0.2M PCA as described previously (12). RNA and DNA were hydrolyzed with 0.6M PCA at 70°C and both radioactivity and DNA content were quantitated in the hydrolyzate. CPM were normalized with respect to DNA content of the hydrolyzates and these ratios (i.e., specific activities) plotted as a function of time. The values indicated are average values for 4-8 animals.
RNA species were made from electrophoretic mobility by the methods of Peacock and Dingman (16,17). Assuming molecular weights of 6-7x105 and 1.5-1.9 x 106 daltons for the 18S and 28S species, respectively, the presumed 45S and 32S species would correspond to molecular weights of 5.0 x 106 and 2.3 x 106 daltons, respectively. Thus a pattern of ribosomal RNA maturation can be seen in the tracheal epithelium which is comparable to that in other types of cells (18)(19)(20)(21). Subsequent studies of RNA metabolism concerned issues related to carcinogenesis in vivo in the respiratory tract of Syrian golden hamsters. First studied was the effect of prior intratracheal. treatment with BP and Fe2O3 in vivo on RNA metabolism in isolated tracheas in vitro. Hamsters received one instillation of either 5 mg BP plus 5 mg Fe2O3 or 5 mg Fe2O3 alone (1) and were killed 48 hr later. Tracheas were isolated MIGRATION DISTANCE,mm FIGURE 2. Electrophoretic distribution of newly synthesized, high molecular weight tracheal RNA as a function of time.
Hamster tracheas were incubated in vitro for 30 min in medium with 20jCi/ml of uridine-3H to pulse label RNA. Whole-cell tracheal RNA was prepared for the epithelium (a) immediately after this 30 min pulse, and after further incubation in unlabeled medium for an additional times of (b) 30 min, (c) 90 min, or (d) 150 min. Hamster liver cytoplasmic RNA labeled with uridine-'4C was added to each sample of tracheal RNA to provide an internal reference for the identification of 18S and 28S RNA species and electrophoresis on acrylamide-agarose composite gels was conducted for 2.5 hr. Gels were cut after electrophoresis, and RNA larger than approximately 7S were included in the analysis. Radioactivity for each gel slice was determined as indicated in the text. Data are presented as the proportion of the total uridine-3H radioactivity present in each gel slice plotted with respect to migration distance.
and incubated with uridine-3H for various times. from the tracheas of hamsters treated with Fe2O3 alone and BP plus Fe2O3 are illustrated in Figure  3. No marked alterations of RNA larger than 28S were evident at 30 min or later. By the time that peaks of 28S and 18S RNA were distinct (60 min and later), however, a consistent alteration in the ratio of 28S to 18S RNA species was evident when tracheal RNA from BP plus Fe2O3-treated hamsters was compared to that of their controls (Fe2O3 alone). Current information suggests that 28S and 18S RNA are formed in equimolar proportions from a 45S precursor rbosomal RNA (22). Based on their equimolar proportions and the estimated molecular weights, the ratios of uridine-3H radioactivity in 28S and 18S RNA should be greater than 2.0 (assuming uridine con-tent to be proportional to molecular weight). This situation is approximated in tracheas of either normal (Fig. 2) or Fe2O3-treated hamsters (Fig.  3). Small deviations from the calculated proportions are presumably a consequence of heterogenous RNA, probably of nuclear origin, which is present throughout the gel. The marked alteration in proportion of 28S and 18S species in tracheas from BP plus Fe2O3-treated hamsters (  (23). Lymphocytes transformed by phytohemagglutinin are the most thoroughly studied of these populations, and in studies with comparable short labeling intervals they show a marked increase in the quantity of 18S RNA as compared to 28S RNA in response to the mitogen (20). This alteration in tracheal RNA metabolism may reflect the stimulation of cell division of tracheal epithelium in response to the BP treatment but not to Fe2O3 alone, which is observed morphologically as a modest hyperplasia during this acute phase after treatment (24).
The observation (25) of an inhibitory effect of vitamin A treatment on experimental respiratory carcinogenesis in vivo was the basis for study of the effect of the in vivo vitamin A status of hamsters on the metabolism of RNA in isolated tracheas. As previously reported (26), an alteration of RNA metabolism was observed in the case of vitamin A deficiency. In this study, isolated tracheas from vitamin A-deficient hamsters and their pair-fed controls were labeled with uridine-3H, and RNA was isolated and separated electrophoretically as usual. In the case of vitamin A deficiency, there was a consistantly smaller proportion of RNA among the high molecular weight, slowly migrating species (Fig. 4; Table 2). This abnormality was reversed following administration of vitamin A in vivo (26). Subsequent studies of RNA metabolism were designed to determine the class of RNA species which were effected in the case of vitamin A deficiency. Specific inhibitors of RNA metabolism were employed to determine whether the decrease in large RNA species related to the heterogenous nuclear RNA or the ribosomal precursor species.
Low concentrations of toyocamycin, an analog of adenosine, selectively inhibited the synthesis of 18S and 28S ribosomal RNA and caused the ac-  cumulation of the 45S ribosomal RNA precursor in L-cells (27,28) and Novikoff hepatoma cells (21). Toyocamycin had the same effect in isolated Environmental Health Perspectives tracheas (Fig. 5a).  The reference point of 30 mm migration has no inherent significance but operationally it permits a numerical comparison of distribution of radioactive RNA in normal and vitamin A-deficient animals (Fig. 4). migrating RNA species in vitamin A deficiency (Fig. 5b) as compared to pulse labeling (Fig. 4). In subsequent experiments, continuous labeling was used to evaluate the effects of inhibitors. The broad peak with 12-18 mm migration is biphasic with continuous labeling. It appears that the more slowly migrating portion of this peak (left half) is the specific region which is decreased in the case of vitamin A deficiency (Fig. 5b). Treatment of tracheas with toyocamycin in vitro did not affect the alteration of RNA pattern in tracheas of vitamin A-deficient hamsters as compared to controls (Fig. 5c). In both cases, however, radioactive RNA accumulated in the region of the biphasic peak presumed to contain the 45S ribosomal RNA precursor (right half).
Further evidence that the right half of the biphasic peak contains the ribosomal precursor comes from an inhibition study using actinomycin D. Synthesis of 18S and 28S RNA was appreciably inhibited in tracheas treated with 0.2 FIGURE 5. Effect of toyocamycin on synthesis of high molecular weight RNA in tracheal epithelium from normal and vitamin A-deficient hamsters. Tracheas were obtained from untreated hamsters, vitamin A-deficient hamsters, or pair-fed vitamin A-normal controls. In cases were toyocamycin was used, tracheas were preincubated in medium containing 1 JAg/ml toyocamycin for 15 min prior to labeling, and toyocamycin at this concentration was also present throughout the subsequent 180 min of incubation. (a) Tracheas were from normal hamsters; labeling with uridine-3H was for 30 min followed by an additional 150 min incubation without ,ug/ml actinomycin D and continuously labeled with uridine-3H (Fig. 6). At 0.1 iAg/ml there was considerably less inhibition and lower doses (e.g., 0.01 or 0.001 ,Ag/ml) did not cause selective inhibition of ribosomal RNA synthesis. Actinomycin D, unlike toyocamycin, prevents the formation of the 45S ribosomal RNA precursor as well as 28S and 18S RNA (29). Consistent with this previous observation, there was a decreased proportion of labeled tracheal RNA in the portion of the gel calculated to contain the 45S precursor in the experiments where actinomycin D was used. The greater inhibition of labeling of 28S RNA than 18S RNA obseved in this instance has also been found in phytohemagglutinin-stimulated lymphocytes (30).  (31,32). This results from a direct interaction with the nucleoplasmic RNA polymerase which is responsible for the synthesis of the DNA-like hetern ogenous nuclear RNA. The electrophoretic pattern of RNA synthesized in isolated tracheas in the presence of a-amanitin had a slightly more prominent 45S region than in untreated controls (Fig. 7). This prominence could be the result of a slightly decreased proportion of synthesis of Effect of a-amanitin on synthesis of high molecular weight RNA in hamster trachea and liver. Liver was obtained from a hamster 2 hr after an intraperitoneal administration of 300 ug/kg of a-amanitin in 0.9°h NaCl solution. Liver was minced finely and incubated in medium containing uridine-3H for 180 min. After incubation the fine pieces of liver were collected and processed identically to tracheal epithelium. Tracheas were from normal hamsters and they were incubated for 180 min in medium containing uridine-3H.
Tracheas treated with a-amanitin were incubated with it at a concentration of 0.2 lAg/ml during the 180 min labeling period only: (o) normal trachea; (A) a-amanitin-treated trachea; (m* ) liver from a hamster treated in vivo with a-amanitin.
species larger than 45S. These large RNA species were most affected by a-amanitin in an experiment -in which hamsters were treated in vivo and uridine-3H labeling of liver followed in vitro (Fig.  7). In contrast to this marked reduction of large RNA, normal liver RN synthesized under comparable conditions has a greater proportion of label in species 45S and larger than does trachea (see Fig. 8). This result suggests that an alteration of metabolism of heterogenous nuclear RNA could produce an electrophoretic pattern which would have a deficit of very large (>45S) RNA as its most obvious effect.
The preceding results suggest that the alteration in RNA metabolism observed in tracheas of vitamin A-deficient hamsters concerns the heterogenous nuclear RNA. Although a complete understanding of this class of RNA has not emerged yet, it is clear that it contains RNA related to messenger RNA (33)(34)(35)(36). The proportion and presumably the types of heterogenous RNA are related to the state of tissue differentiation. The greatest differences between electrophoretic patterns of RNA synthesized in trachea, liver, and esophagus in vitro were in species 45S and larger (Fig.8). The differentiation of the tracheal epithelium of a vitamin A-deficient hamster is morphologically changed from that of the normal trachea. . Electrophoretic distribution of newly synthesized high molecular weight RNA from hamster trachea, liver, and esophagus. Tissues were obtained from normal hamsters and all were incubating for 180 min in medium with uridine-3H. Tracheas were prepared and labeled as described previously. Liver from a normal hamster was minced, labeled, and analyzed as described in the legend of Fig. 7. The esophagus of a normal hamster was washed extensively with phosphate buffered saline then cut open. It was incubated with medium containing uridine-3H as a flat sheet comparable to an isolated trachea. Studies concerning binding of BP to DNA in isolated tracheas form a second major area of our biochemical studies of tracheobronchial epithelium. These observations have been published recently (38)(39)(40), and the following is limited to an abbreviated reiteration of these results.
The motivation for these studies was also to learn more about factors related to the induction of respiratory tract tumors in vivo in life-time carcinogenesis studies with Syrian golden ham-sters. Initial experiments were designed to demonstrate purification and banding of DNA from tracheas, and to determine whether binding of carcinogens during in vitro incubation reached detectable levels. Tracheas were obtained from normal hamsters and from hamsters which received an intratracheal administration of BP plus Fe2O3 or Fe2O3 alone, 48 hr prior to sacrifice. The tracheas were incubated in vitro in medium containing BP-3H and following incubation the epithelium was removed for further study. Unincorporated BP-3H was carefully extracted from the partially purified nucleic acids. RNA and proteins were enzymatically digested, leaving a DNA preparation which banded as a sharp peak in CsCl gradients with low background absorbance at 254 nm (see examples in Fig. 9). Radioactivity was determined for all gradient fractions, and ultraviolet absorption was carefully quantitated for peak fractions. A greater quantity of radioactivity banded coincidently with the DNA peak from the tracheas of hamsters previously treated with BP plus Fe2O3 in vivo than in untreated hamsters. In contrast, an intratracheal dose of Fe2O3 alone did not induce greater binding of BP-3H. These results are presented quantitatively in Table 3. Here, the specific activity of binding is determined as the quantity of radioactivity from bound BP-3H (DPM) normalized to the DNA content (,ug DNA) of the peak fractions.
The next experiments were to demonstrate that the increased BP binding in tracheas of hamsters treated with BP plus Fe203 in vivo was the result of activation of BP in vitro rather than a nonspecific phenomenon. Tracheas were obtained from animals treated with BP plus Fe2O3 48 hr earlier. Tracheas incubated at 0°C or with 7,8-benzoflavone (BF), an inhibitor of BP metabolism (41), added to the medium were compared to tracheas incubated under standard conditions. The results of these studies (Table 3) indicate that both BF and 0°C incubation inhibit binding of BP-3H to DNA in isolated tracheas. These results suggest that the observed BP binding of isolated tracheas is the summation of two processes. Binding is low in tracheas of normal hamster or in tracheas treated with BF, but somewhat higher than that observed after incubation at 0°C. Prior treatment with BP plus Fe2O3 in vivo induces a much higher level of binding in isolated tracheas, and this induced binding is reduced to normal levels by addition of BF to the incubation medium.
Since initial studies demonstrated that BP binding to tracheal DNA was detectable and the August 1976 result of both basal and inducible components, the objective of these studies became the evaluation of these properties for their correlation with susceptibility to carcinogenesis in vivo. It was hoped that, in addition to evaluating genetic metabolic capacity, which previous reports related to susceptibility in man (42) and mouse (43), these properties could evaluate other biological properties in the target organ which serve as cofactors in determining susceptibility. Vitamin A deficiency was one such biologic factor of interest since marginal vitamin A status has been shown to increase the incidence of aflatoxin-induced colonic carcinomas in rats (44). Vitamin A-deficient and pair-fed, vitamin A-normal hamsters were prepared as described previously and the binding of BP-3H to DNA in isolated tracheas determined following incubation in vitro. The BP binding level in the tracheas from vitamin A-deficient hamsters was nearly four times greater than in the vitamin A-normal controls. This elevated binding level was also inhibited by incubation with BF in the medium or by incubation at a temperature of 0°C (Fig.9). Addition of 5-15 *g/ml of retinyl acetate to the medium in which tracheas from vitamin A-deficient hamsters were incubated caused substantial inhibition of BP binding, but less than that caused by BF. In other studies two inbred strains of hamsters were selected for their demonstrated susceptibility to subcutaneous and gastrointestinal carcinogenesis (45). Levels of binding of BP in vitro in both inbred strains were about twice those of random bred hamsters following treatment with BP plus Fe2O3 in vivo. In one additional experiment, levels of BP binding were evaluated in random bred hamasters of 4, 8, and 12 weeks of age. There was substantially greater binding in the tracheas from the two younger groups. The significance of these observations will not be clear until we know the relationship of respiratory tumor incidence to vitamin A-deficiency, inbred hamster strain, and age at the time of carcinogen treatment. For the moment, however, the observation of high levels of binding in biologic states which appear likely to be states of increased susceptibility suggests that this type of binding assay may hold promise of distinguishing both genetic and acquired susceptibility factors which predispose to the development of lung cancer in the hamster model.

Conclusion
These results illustrate that biochemical studies on a respiratory epithelium, from hamster trachea in this case, can reach substantial sophistication. Methods now exist for a variety of biochemical investigations on this tissue, and the accomplished studies, hopefully, will give encouragement to further exploration of the biochemical properties of this target tissue.
Environmental Health Perspectives RNA metabolism in the epithelium of isolated tracheas from normal hamsters shares a number of characteristics previously reported for other tissues and cell types. Alterations in the electrophoretic distribution of newly synthesized RNA were observed for two abnormal states in vivo: the acute phase following intratracheal treatment with BP plus Fe2O3, and vitamin A deficiency. Efforts to distinguish the class of RNA decreased in vitamin A deficiency have provided information on the effects of three inhibitors of RNA metabolism which illustrate a variety of acute states of biochemical toxicity in this target tissue. Comparable studies of the effects of vitamin A deficiency on RNA metabolism might not have been possible in many other tissues, or tissue culture cells, whose differentiation is not vitamin A dependent. In the case of BP binding studies, it was useful to learn that BP is metabolized in the tracheal epithelium and that it binds to DNA. This could have been anticipated, however, from the fact that instillation of BP into the respiratory tract can cause tracheal tumors. Only by direct studies of this epithelium, however, could we learn of the capacity of this tissue to undergo induction of a mechanism which increases BP binding. Furthermore, the evaluation of other biological states in vivo (e.g., vitamin A deficiency, genetic strain) for their effects on BP binding,at this target site could only be studied in a respiratory epithelium. It is clear, therefore, that certain kinds of questions can only be studied, or studied best, in the specific target tissue. This belief forms the basis for initial studies of human respiratory epithelium (46) in an effort to identify specific biochemical characteristics of this human target tissue.
The ultimate importance of biochemical studies thus goes beyond comparison with other tissues. The human population shows a diversity of inherited and acquired characteristics, and this may explain why at a given level of injurious exposure, clinical cancer or emphysema, etc. develops in some individuals and not in others. It is to be hoped that biochemical and physiological studies of the respiratory tract in experimental animals can offer some insight into the genetic properties and acquired conditions which make one individual susceptible an'd another refractory to a given exposure. By specifically protecting the susceptible, perhaps we can more finely adjust the scales of societal benefit versus individual risks to maximize the advantages for all.