Cytotoxic effects of low levels of 3H-, 14C-, and 35S-labeled amino acids.

Tissue injury by radiolabeled amino acids may severely affect experimental results. In this report, 3H-, 14C-, and/or 35S-labeled proline, serine, lysine, and methionine at concentrations of 1 and 5 microCi/ml were shown to cause severe injury in organ cultures of embryonic rat lungs. This injury was evident by 6 h and was amplified by 4 days of culture. This injury was characterized with light and electron microscopy, with morphometric analysis of growth, and with quantitation of the total protein and DNA/lung. After 6 h with 5 microCi/ml of 14C- or 35S-amino-acid there were more signs of cell degeneration, and by 24 h the labeled lungs were smaller, showed more signs of cell degeneration and death, and contained 30 to 60% less new protein and DNA than control lungs. After 24 h with 5 microCi/ml of 14C- or 35S-amino-acid the total protein and DNA/lung began to decrease. This toxicity was directly proportional to the amount of intracellular decay of each isotope. With 14C- and 35S-amino-acids, lung growth slowed with approximately 100 disintegrations/cell/day (d/c/d), growth stopped with approximately 200 d/c/d, and atrophy occurred with approximately 300 d/c/d. Cell proliferation, cell differentiation, and bronchial branching continued through 4 days even though atrophy occurred with greater than 200 d/c/d. With 3H-amino-acids, growth slowed with approximately 200 d/c/d and stopped with approximately 400 d/c/d. However, no toxicity was evident with less than 60 d/c/d of 14C or 35S, or with less than 90 d/c/d of 3H. These data suggest that the amounts of intracellular decay of these weak beta-emitting isotopes should be strictly limited. Increasing amounts of tissue injury occurred with 14C or 35S at greater than 10,000 dpm/micrograms of DNA, and with 3H at greater than 20,000 dpm/micrograms of DNA.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. disintegration of 3H (4). This damage of DNA results in a slowing or block of cell proliferation (4)(5)(6)(7)(8)(9)(10). For example, with cultures of HeLa cells, 0.2 pCi/ml of ["Hlthymidine reduced cell proliferation by -10% and 0.4 pCi/ml reduced cell proliferation by -99% (9). Similarly, with cultures of WI-38 cells, a 0.1 pCi/ml concentration of [3H]thymidine slowed cell proliferation by -50% and 0.25 pCi/ml stopped cell proliferation completely (10). This block of cell proliferation is most efficient with 3H in DNA, less efficient with "H in nuclear RNA, and least efficient with "-labeled amino acids (11). This is due to the fact that the path length of 90% of the ,8-electrons of 3H decay is less than 1 pm (12). Nevertheless, cell proliferation is blocked by the incorporation of ~3H]a-aminoisobutyric acid, which is transported into the cytosol as a small neutral amino acid but is not incorporated into cellular components (13). This phenomenon is referred to as reproductive cell suicide.
A review of numerous radiotracer studies of protein synthesis and other metabolic processes shows that many investigators routinely use high concentrations of radiolabeled amino acids for periods from 6 h to a few days. Furthermore, the corresponding unlabeled amino acid is often omitted from the culture medium. It appears that many investigators are not aware that 3H-, I4C-, and "S-amino-acids quickly result in the type of tissue injury that is described in this report. For example, we were quite surprised when 5 pCi/ml of CHIand ['4C]proline were severely toxic in cultures of embryonic chick somite mesoderm' and in cultures of both embryonic rat and chick lungs (3).
Since radiolabeled amino acids are widely used at toxic concentrations it is important to show how trace levels of these labeled precursors can affect tissue function. Therefore, these studies were designed to learn 1) whether the inhibition of tissue growth is due to the radiolabel or a toxic contaminant in the labeled amino acid preparation, 2) whether different types of labeled amino acids are equally toxic, 3) whether the toxicity is reduced by dilution with increasing concentrations of L-amino acids in the nutrient medium, or 4) whether the toxicity is proportional to the transition energy of the isotope, to the concentration of radioactivity in the nutrient medium, to the amount of labeled amino acid that is incorporated, or to the function of the isotopic atom or it's transition product.

MATERIALS AND METHODS
Lung Bud Isolation and Culture-Timed pregnant, 200 g, BLU Sprague-Dawley rats were obtained from Blue Spruce Farms, Altamont, NY. At either 13.5 or 14.5 days of pregnancy, the rats were killed by decapitation and the uterus was removed and placed in Simm's balanced salt solution (14). The embryos were removed from the uterus, the trachea and lung buds were cleanly isolated, the main ' Unpublished experiments.  15). This F-12C was prepared by Grand Island Biological Co. without proline, lysine, valine, or glutamine (formula No. 78-0420). When fresh media was prepared these amino acids were added to a final concentration of 2.4 mM L-proline, 1.6 mru L-lysine, 0.8 mM L-valine, and 8.0 mru L-glutamine (Table I). To increase the specific activity of radiolabeledproline or lysine in selected experiments the concentrations of proline and lysine were reduced to 0.8 mM. This modified nutrient mixture was referred to as F-12mC (Table I). In selected experiments proline was omitted from the F-12C and this mixture was referred to as F-12C(pro). During the course of this study it was found that the 1% nutrient agar substrate was much easier to prepare and results were more consistent with a mixture of 20% SBSS, 10% FBS, and 70% F-12C. This mixture was used for subsequent experiments. In all experiments, the nutrient agar and liquid nutrient feeding media were prepared as described previously (16).   I  Amino  acid content  of HamS F-12, F-12C, and F-12mC,  and the  estimated  concentration  of amino acids in fetal bovine serum   F-12"  F-12@  F-12mC FBS' - At the end of 1, 2, and 4 days of culture, the lungs on one dish from each 0, 1, and 5 @Zi set were photographed and collected for analysis of the total protein, total DNA, and t,otal dpm of incorporated radioactivity/lung. To further characterize growth, a Digital Equipment GT-40 computer with a Summagraphics tablet was used to measure the area of each lung bud in photomicrographs of living cultures.
In some experiments one pair of lungs from each dish was also fixed for scanning and transmission electron microscopy after 6 and 12 h and at daily intervals thereafter. For chemical analysis the lungs from each dish were washed for 2 min in 30 ml of cold SBSS, boiled 3 min in 0.5 ml of water containing 0.05 pm01 of N-ethylmaleimide, and stored at -20 "C. Since the right lung bud was always 1.5-2 times larger than the left lung (Fig. l)   trichloroacetic acid was added to a concentration of 5%. The sample was incubated 2 h at 4 OC, and centrifuged at 2000 X g for 30 min at 4 "C. The supernatant was removed, the pellet was dispersed by sonication in 0.5 ml of H20, and 100-pl aliquots of both the supernatant and suspended pellet were counted in Liquiscent. These analyses showed that with each labeled precursor -90% of the radioactivity in the washed lungs was precipitated by 5% cold trichloroacetic acid. The changes in average area correspond directly with the changes in total protein and DNA per lung ( Table 11).

--
The disintegrations per cell per day (d/c/d) were estimated with the formula: dpm/lung X (pg of DNA/lung)" X 1440 min/day X (143,000 cells/pg of DNA)" = estimated d/c/d. The d/c/d were then used with the following formula from Casarett (23) to estimate the rads/cell/day.  6.88 f 0.6 9.9 7.55 f 1.1 9.2 1.38 f 0.5 13.5 Estimated rads/cell/day The average transition energy per disintegration ( E ) for 3H is 0.0055 MeV, and for I4C E is 0.050 MeV (17). The estimate of 7 X 10"" g/cell was obtained with the following formula: 10 X g of protein/lung X (estimated number of cells/lung)" = estimated g/cell. This assumes that the average cell is -10% protein. These values are obviously only rough estimates but they facilitated the comparison of the toxic levels of radioactivity in the cultured lung buds with data from other experimental systems.
Two methods of estimating cell volume were used to estimate the concentrations of incorporated isotope in lung cells. One method assumed that the average cell was a 10-pm sphere and this gave an average cell volume of 5.236 X 10"" ml. Comparable values were obtained if one assumed that the average cell in these lung buds was 10% protein. The average protein:DNA ratio was -10 (Tables 111-VI) and the average diploid rat nucleus contains -7 pg of DNA (8). Therefore, the average cell was estimated to contain -70 pg of protein and have an average cell weight of 700 pg. With a specific density of 1.10 this weight would occupy a volume of 6.36 X 10"" ml . Thus, the average cell was estimated to have a volume of 5.8 X 10"' ml and this value was used to estimate the intracellular concentration of each isotope.
The amount of radioactivity remaining in the nutrient agar was also determined by counting aliquots of the nutrieilt agar that was  Table 111. dissolved in 3 ml of warmed water. These analyses showed that 5 to 10% of the radioactivity in each dish was incorporated by five pairs of 14.5-day-old rat embryo lung buds.
For microscopic analysis the lungs were fixed for 24 h at 4 "C in 4% glutaraldehyde in 0.1 mM cacodylate buffer at pH 7.4, and post-fiied for 1 h at 4 "C in 1% osmium tetroxide in the same buffer. In some experiments the tissues were stained en bloc with ruthenium red as described previously (24). The fixed lungs were stained en bloc with uranyl acetate (24), and were cut transversely into three pieces that included the proximal, intermediate, and distal airways, respectively. These pieces were dehydrated in ethanol, followed by propylene oxide, and were embedded in Epon:Araldite. One micron sections of lungs were deplasticized and stained as described previously (24). Thin sections, stained with uranyl acetate and lead citrate, were examined with a Phillips 301 electron microscope.

Effects of [%']Proline
Lung growth, bronchial morphogenesis, and cytodifferentiation were identical in these studies with that described previously (3). The original goal of these experiments was to study the synthesis and deposition of collagen during the period of bronchial morphogenesis. Surprisingly, there was   (-pro) nutrient media. After 12 h of culture, the explants with ['"Cl proline appeared darker and more granular than control explants. This density and granularity corresponded to an increase in cell death and an accumulation of debris filed secondary lysosomes as described below. At 24 h this difference in translucency and granularity was more pronounced and the labeled lungs were smaller than control lungs (Figs. 1 and 2). These differences were most pronounced in cultures of 13.5-day-old lungs ( Fig. 2). This difference in size was accentuated after the 2nd day of culture when the labeled lungs began to atrophy while the control lungs continued to grow ( Figs. 1 and 2). Thus, bronchial branching continued but growth stopped and atrophy occurred in the presence of 5 @i of [ 'Clproline.
The visible differences in living cultures consistently corresponded to a marked decrease in the rate of accumulation of both protein and DNA in explants growing in the presence of ["Clproline (Tables III and IV (Table IV). Consequently, these results indicated that the ['4C]proline itself rather than a toxic contaminant was responsible for the inhibition of lung growth.  Table V). was assumed to be 0.374 mM in the FBS and it was 2.4 mM in the F-12C and 0.8 mM in the F-12mC (Table I and  As shown in Table V the toxicity of the [14C]proline was still present but there was less inhibition of growth as the incorporation of [I4C]proline was reduced by dilution with the increased amounts of L-proline in the F-12C (cfi Tables I11  and V). With a 2.2 X dilution of 5 pCi of [%]proline, atrophy occurred after culture day 1 (Table 111), but atrophy did not occur until the 3rd or 4th day of culture with 5 pCi/ml of ["C] proline diluted 100 X with 70% F-12C (Table V). The inhibition of growth in cultures with F-12mC was intermediate between that with F-lBC(-pro) and F-12C (data not shown). Thus, the inhibition of growth varied inversely with dilution and directly with the amount of incorporation of [I4C]proline.

Effects of ['4C/Serine
Once it was clear that the incorporation of [I4C]proline was the cause of the inhibition of lung growth, it became important to know whether other 14C-amino-acids have similar effects. To answer this question 14.5-day-old lung buds were cultured with either 0, 1, or 5 pCi of [14C]serine with a specific activity of 159 or 178 mCi/mmol. This experiment was repeated three times and in each case the inhibition of lung growth was very severe with 5 pCi/ml and moderately severe with 1 (Table VI). This indicated that during a 24-h pulse with 1 pCi of [I4C]serine there was 33% and with 5 pCi there was 50% less new protein accumulation than in lungs growing without a labeled amino acid. By the end of 4 days of culture, the total protein/lung had increased approximately 300% in the control cultures and approximately 100% in cultures containing 1 pCi/ml. However, with 5 pWml the total protein and DNA decreased after the 1st day of culture, and with 1 pCi/ml there was no increase in DNA after day 1 (Table VI). Nevertheless, the dpm of I4C/ pg of DNA continued to increase (Table VI), suggesting that protein synthesis continued through 4 days of culture. Table VI there was approximately 58 d/c/d after 24 h of culture with 1 pCi/ml of [l4C]serine. At 48 h this value had increased to -113 d/c/d, and by this time there was a notable decrease in lung growth. With 5 pCi/ml there were -213 d/c/d at 24 h, and during the 2nd day of culture lung growth stopped and atrophy began. This indicated that the rate of cell death exceeded the rate of cell proliferation when the rate of intracellular decay of 14C was >200 d/c/d or -20,000 dpm/pg of DNA. This is equivalent to -230 rads/ cell/day.

Lung growth with or without [14C]serine
Microscopic studies showed that there was a consistent increase in the signs of cell degeneration and death in labeled cultures (Figs. 6 to 8). After 6 h of culture with 5 pCi/ml of [I4C]serine, increased numbers of secondary lysosomes or digestive vacuoles had begun to accumulate in both epithelial and mesenchymal cells (Fig. 6). By 24 h large complexes of digestive vacuoles were present in many mesenchymal cells, and pycnotic cells and cell debris were present in the blood islands (Fig. 6). In addition, an extensive accumulation and hydrolysis of glycogen resulted in the appearance of localized cytoplasmic vacuolation in nearly all epithelial cells in labeled lungs (Figs. 7 and 8). This increased accumulation of digestive vacuoles and cell debris in the mesenchyme and the accumulation of glycogen in the epithelium continued through 4 days of culture (Figs. 7 and 8). These signs of cell degeneration and death were directly related to the increased density and granularity of living cultures, and to the slowing of growth and onset of atrophy in labeled lungs.
Mitotic figures were present in both epithelial and mesenchymal cells in both labeled and unlabeled cultures throughout 4 days of culture (Figs. 7 and 8). Thus, cell proliferation and bronchial branching both continued through 4 days in the presence of 5 pCi/ml of [I4C]serine.

Effects of ['4C]Lysine
These experiments were designed to ask whether [I4C]lysine is more toxic than [I4C]proline or serine. Since lysine constitutes 17% of the residues in histone proteins, the proportion of intranuclear decay should be increased with this label. The fact that it is both cationic and an essential amino acid might also affect the toxicity of [14C]lysine. Three sets of dishes These results were inconsistent with the hypothesis that the toxicity would be increased by enrichment of 14C-labeling of nucleoproteins.

["C]versus ["%]Methionine
[I4C]Methionine-1n three identical experiments, 8 or 10 lung buds were grown on 1 ml of agar in a 101080 mixture of SBSS:FBS:F-12C containing either 0, 1, or 5 pCi of ["C] methionine. The results of these experiments were essentially identical with those with other 14C-amino-acids (cf Tables 111,  VI, VII, and IX). In each case 5 pCi/ml were severely toxic and 1 pCi/ml was moderately toxic (Fig. 10 and Table IX). After 12 h the lungs with 5 pCi/ml were more dense and granular, and by 24 h the labeled lungs were smaller and contained 20 to 25% less protein and DNA than control lungs. containing 0, 1, or 5 pCi/ml of ['4C]lysine with a specific activity of 338 mCi/mmol were used in each experiment. In two experiments with 70% F-12mC and 10% FBS, the final concentration of L-lysine was -0.5905 pmol/ml, and in two experiments with 80% F12C and 10% FBS the final concentration of L-lysine was -1.3105 pmol/ml (Table I). The results of these experiments showed that [I4C]lysine also inhibited lung growth (Fig. 9). This inhibition of growth was greatest with 5 pCi/ml, and it decreased as the dilution with L-lysine increased (cf Tables VI1 and VIII). This inhibition of growth with 5 pCi/ml of [I4C]lysine was less than that with comparable levels of incorporation of [I4C]serine ( c f Tables VI and VIII).  nification: A, 300 X; B, 1100  The microscopic differences in these lungs were identical with those with 5 &i/ml of ['?]serine. There was a rapid increase in the number of debris-filled secondary lysosomes in mesenchymal cells, dead cells, and cell debris in blood islands and glycogen accumulation in bronchial epithelial cells (data not shown). These differences continued to increase through culture day 4. After 24 h atrophy was evidenced by a decrease in size (Fig. 10) and by a decrease in the total protein and DNA/ lung (Table IX). However, bronchial branching and cell differentiation continued, and mitotic figures were present in all cultures. This showed that the toxicity of ['4C]methionine was indistinguishable from that with other "C-amino-acids and this suggested that the toxicity was not related to the function of the labeled amino acid. AS shown in Fig. 10 and  (Table IX). These d/c/d were estimated to represent doses of 70, 114, and 458 rads/cell/day.
[35S/Methionine-0ne of the major purposes of these studies was to learn whether the number of heavy atoms incorporated or the position or function of the isotopic atom(s) affected the toxicity of the radiolabeled amino acids. To help answer this question the previous experiment was repeated four times with 0, 1, and 5 @i/ml of ["5S]methionine.  (Table I). Each microcurie aliquot contained 4.06 X 10" labeled molecules or 6.74 x 10m7 pmol of [35S]methionine (Table II). This gave a minimum final dilution of 257,600 with 1 @i/ml and 51,500 with 5 &i/ ml (Table X). These dilutions of the ["5S]methionine were -4,500 times that of the [14C]methionine in the previous experiments (cf. Tables IX and X). The results of these experiments showed that in spite of the large dilution there was a severe inhibition of lung growth in all cultures with 5 @/ml and a moderate inhibition in those with 1 &i/ml ( Fig. 11 and Table X). At 12 h the labeled lungs were more dense and granular, and at 24 h the labeled lungs were smaller and contained -60% less new protein than control lungs. After 24 h with 5 &i/ml the lungs began to atrophy but bronchial branching continued through 4 days. As in previous experiments the atrophy was evidenced by a progressive decrease in the average area, total protein, and DNA/ lung ( Fig. 11 and Table X).
Light and electron microscopic examination showed that the morphologic differences between labeled and control lungs were identical with those seen previously in experiments with [14C]serine (Figs. 6 to 8). The bronchial epithelial cells accumulated large amounts of glycogen in their cytosol, debrisfilled secondary lysosomes accumulated in mesenchymal cells and dead cells, and cell fragments accumulated in blood islands. After 4 days of culture, type II pneumocytes begin to differentiate in all cultures but there were fewer lamellar bodies in the glycogen laden epithelial cells in cultures with 5 @i/ml.
Mitotic figures were also present in both the epithelium and mesenchyme and there were many morphologically normal cells in all cultures. This showed that there was much more evidence of cell degeneration and death, but there was not a generalized killing of cells in the labeled lungs (data not shown).
The toxic effects of [35S]methionine were equal to or greater than those of [14C]methionine (ct Tables IX and X). With both labels the rate of accumulation of protein and DNA was markedly slowed by 1 @/ml and there was a net 10~s of protein and DNA after a short period of growth with 5 &i/ ml. Furthermore, the amount of radioactivity per lung with [35S]methionine was equal to or slightly greater than that with ['4C]methionine (Tables IX and X). This equivalence of incorporated radioactivity resulted from the fact that the decay constant of 35S is -2.3 x lo4 larger than that of 14C. Even showed that the inhibition of growth was more directly related

Lung growth with or without [I4C]lysine at higher levels of dilution with L-lysine
The values represent the mean and standard deviations obtained in two experiments with cultures of 14.5-day-old embryo lung buds that were grown for 1, 2, or 4 days on 1% agar in a 2010:70 mixture of SBSS:FBS:F-12C containing either 0, 1, or 5 pCi/ml of [I4C]lysine with a specific activity of 338 mCi/mmol. that showed that growth slowed with -100 d/c/d, and atrophy began with >200 d/c/d of incorporated 14C (cf Tables IX and  X). This suggested that the toxic effects of these compounds was not directly related to the position, function, or transmutation of the isotopic atoms in the labeled amino acids.

["C]versus ['HIProline
In the initial experiments with tritiated amino acids the effects of [3H]proline were compared to those of [14C]proline. Five sets of duplicate dishes of 10 lung buds were used with a 10:10:80 mixture of SBSS:FBS:F-l2C(-pro). Three sets of duplicate dishes contained either 0, 2.5, or 5 pCi/ml of ['HI proline (NET 323) with a specific activity of 23 Ci/mmol, and two sets of dishes contained either 1 or 2.5 pCi/ml of ["C] proline (NEC 285) with a specific activity of 255 mCi/mmol. After 4 days of culture the lungs were assayed for total protein and DNA. As shown in Fig. 12 and Table XI of [I4C]proline and 58% with 5 pCi/ml of ["Hlproline (Table  XI). At equal concentrations of radioactivity (2.5 pCi/ml), there was -60% less inhibition of lung growth with ["Hlproline than with [14C]proline (Table XI). This suggested that 3H with an average transition energy of 0.0055 MeV was less toxic than C with an average transition energy of 0.05 MeV/disintegration. This suggestion was complicated, however, by the differences in dilution that resulted from the differences in specific activity, by the differences in the decay constants of 3H and I4C, and by the differences in the number of heavy atoms/ molecule in these labeled precursors. For example, at a concentration of 2.5 pCi/ml there were 78.4 x more molecules of [14C]proline than [3H]proline, and -455 atoms of I4C for each atom of 3H. This suggested that the greater dilution of pH] proline contributed to the decrease in the inhibition'of lung growth with this precursor. Furthermore, with 1 pCi/ml of ['4C]proline there were -47,000 dpm or -2,350 MeV/min/ lung and with 5 pCi/ml of ['Hlproline there were -124,000 dpm or -682 MeV/min/lung (Table XI). Since 682 MeV of 'H was more toxic this suggested that the ionizing energy of 'H was considerably more toxic than that of 14C. This may be due to the fact that most of the energy of ' H decay would be absorbed within the cell of origin while much of the energy of 14C decay would be released outside of the cell of origin.
In the next experiments in this series the effects of dilution of ['Hlproline were studied by using 1 and 5 pCi/mol of ['HI proline with nutrient medium containing either F-12C, F-12mC, or F-lBC(-pro) ( Table I). This gave final dilutions of 57,000, 19,700, and 112 with 1 pCi/ml and dilutions of 11,400, 3,900, and 22 with 5 pCi/ml. The results of these experiments showed that lung growth was significantly improved by increasing dilutions of 5 pCi/ml of ['Hlproline (data not shown).  Table VIII. In 4-day cultures of 14.5-day-old embryo lungs, the total protein/lung increased 300% in lungs grown without ['Hlproline, and 222% in lungs grown with 5 pCi/ml of ['Hlproline that was diluted 11,400 X with L-proline in the nutrient medium. Similar results were obtained with duplicate cultures of 13.5-day-old lung buds (data not shown), and this suggested that lung growth was only reduced by about one-third when 5 pCi/ml of ['Hlproline is diluted by large amounts of Lproline. However, when the intracellular decay of "H reached 300 d/c/d growth stopped (Table XII). This is equivalent to -30,000 dpm/pg of DNA or -38 rads/cell/day.

Effects of ["HISerine
The next series of experiments was designed to determine whether other 'H-labeled amino acids also inhibit tissue growth. In this series, three sets of dishes containing 0, 1, or 5 pCi/ml of ['Hlserine with a specific activity of 16 Ci/mmol were used with a 2010:70 mixture of SBSS:FBS:FlBC. As shown in Fig. 13 and Table XI11 these experiments showed that the inhibition of growth with 5 pCi/ml of ["Hlserine was comparable to that with 5 pCi/ml of ["Hlproline. Furthermore, with lpCi/ml of ['Hlserine at a dilution of 10,790 there was   Fig. 10.    Fig. 11 Table X. little inhibition of lung growth during 4 days of culture (Fig.  13). DISCUSSION The observation that the toxic effects of 5 pCi/ml of I4Cand 35S-labeled amino acids were evident after only 6 h of culture suggests 1) that the injury resulting from the intracellular decay of these isotopes is more severe than was previously anticipated, 2) that the injury occurs during the first few hours of labeling and is amplified by 4 days of culture, and 3) that the injury occurs before a block of cell proliferation would be evident. These results also indicate that the time and amount of exposure of cells to the intracellular decay of these isotopes should be strictly limited.

Label
Atrophy is a decrease in tissue mass and it is the product of a negative anabolic-catabolic balance.
The atrophy in the labeled lungs may have resulted from a decrease in cell proliferation and protein synthesis and/or from an increase in protein degradation and cell death. There was no evidence in these studies to quantitate the contribution of each of these processes but a number of observations suggest that an increase in protein degradation and cell death may have been most important. For example, microscopy consistently showed that there were many more cells containing debris-fied secondary lysosome, there were more dead cells, and atrophy occurred in the labeled cultures. At the same time bronchial branching continued, mitotic figures were consistently present, and the dpm/pg of DNA continued to increase, suggesting that cell proliferation and protein synthesis continued through 4 days of culture. Furthermore, these results suggest that if the decay of 3H, 14C, and 35S accelerates protein degradation, radiotracer studies of protein turnover should be carefully controlled.
The consistency of these results suggests that this may be a general toxic effect of radiolabeled amino acids. The data also suggest that the toxicity is directly related to the amount of incorporation of radiolabel. The toxicity was proportional to the concentration of radiolabel and inversely proportional to the amount of dilution of labeled precursors by unlabeled amino acid in the nutrient medium. The specific involvement of the radiolabel is also strongly supported by the results of experiments comparing the effects of [''Clproline and ["C] proline that were chromatographed on the same system ( Fig.  3 and Table IV). The 14C was the only variable and the lung buds consistently grew well in the presence of the ["C]proline samples and failed to grow in the presence of [14C]proline.
These results also suggest that there may be a "safe dose" of intracellular decay of these isotopes. No signs of injury were evident with t60 d/c/d of 14C or 35S or with <90 d/c/d of 3H. These values were estimated to represent an absorbed dose of -69 rads/cell/day from 14C or 35S, or 11.3 rads/cell/ day from 3H. In contrast, growth slowed with -100 d/c/d, growth stopped with -200 d/c/d, and atrophy occurred with >300 d/c/d of 14C or 35S. This indicated that the rate of cell death equaled or exceeded the rate of cell proliferation when there were >200 d/c/d of 14C or 35S or >400 d/c/d of "H. With embryonic rat tissues 60 d/c/d equals -6,000 dpm/pg of DNA while 100 d/c/d is equivalent to -10,000 dpm/pg. Therefore, with these embryonic lung tissues t6,000 dpm of intracellular decay/pg of DNA might be considered as a "safe dose" while >10,000 dpm/pg of DNA would severely injure tissues during the first 12 h of labeling. It will now be important to learn whether similar concentrations of intracellular decay of these isotopes have similar effects on other cell populations.
As noted above, the average cell in the lung buds was estimated to occupy a volume of 5.8 X 10"" ml. With 60 d/c/ d this volume would contain -1.9 X lo-' pCi or -32 pCi/ml.
With 200 d/c/d the intracellular concentrations would be -108 pCi/ml. These values are respectively 3,200 X and 10,800 X the maximum permissible concentration of I4C in water for man (  17). However, these concentrations are less than those achieved in many experimental studies of the synthesis or degradation of tissue components.
It is not clear whether the toxicity of any of these precursors was selective for different cell populations but the signs of injury were distinctly different in epithelial, mesenchymal, and hematopoietic cells in the cultured lung buds. There was an excessive accumulation of glycogen and hydropic degeneration in epithelial cells, accumulation of secondary lysosomes and residual bodies in mesenchymal cells, and a condensation and fragmentation of hematopoietic cells in labeled lungs (Figs. 6 to 8). These differences suggest that additional studies should be done to determine whether this injury results in a change in the relative proportions of different types of cells, or in the relative proportions of different proteins or other cell products that accumulate in labeled and unlabeled tissues.
Toxic effects of short exposures to radiolabeled amino acids may be masked by a number of variables of tissue isolation and culture. These variables include 1) differences in the amounts of tissue injury resulting from the procedures of tissue isolation and explanation, 2) variations in the age, stage of development, or functional state of the tissues, 3) variations in the amounts of protein or DNA per unit mass of freshly isolated tissue, or 4) the presence of other toxic or deleterious factors in the culture system. In this study the effects of these variables were minimized by the fact that whole lung buds could be isolated and cultured without exposure to proteolytic enzymes or the trauma of slicing or mincing. Furthermore, many variables in the culture system were tested to select favorable conditions for the maintenance of tissue growth and differentiation and to minimize cell degeneration and death (3). Finally, since the presence or absence of the radiolabeled amino acid was the only variable in each experiment, the reduction of growth in response to this variable was easily quantitated. This may answer the question of why the toxicity that was so obvious in this report has not been described previously.
The results described in this report have serious implications for experimental systems using >1 pCi/ml concentrations of labeled amino acids to study the synthesis and turnover of protein. Biologists may be injuring tissues rather severely in radiolabeling experiments, even though the tissues continue to incorporate radiolabeled precursors and undergo cell differentiation and morphogenesis.