Rate of incorporation of [3H]thymidine in various tissues of the mouse. A basis for the evaluation of genotoxic effects of chemicals

[3H]Thymidine has been extensively used as a selective precursor to DNA in studies on the kinetics of cell proliferation. We have become interested in measuring early inhibition of the DNA synthesis in various organs of intact animals for detecting genotoxic properties of chemicals. Such experiments should, for convenience and to achieve a large capacity, be performed in the simplest way possible. The present paper deals with some practical aspects on the use of [3H]thymidine in vivo. [6-3H]Thymidine was injected intraperitoneally in mice and the uptake of radioactivity was evaluated by using whole-body autoradiography and liquid scintillation spectrometry. Autoradiograms of sections washed with trichloroacetic acid and methanol were compared with those subjected only to freeze-drying. Liquid scintillation counting was performed of total, non-volatile, acid-insoluble and DNA-associated radioactivities. A rapid increase of the [3H]thymidine incorporation was seen during the first hour after the injection. Further prolongation of the survival time did not result in any significant increase of the incorporated radioactivity. Moreover, there were only slight differences between the autoradiograms from extracted and non-extracted sections. Radioactivities associated with DNA closely correlated to those representing acid-insoluble material, indicating that acid-insoluble radioactivity provides a good estimate of the [3H]thymidine incorporation into DNA.     

Development and Localization of the Thymidine Phosphorylating Systems in Brain

Abstract: The development of the thymidine phosphorylating systems was studied in various regions of brain. Brain slices from cerebellum, brain stem, and forebrain of rabbits 2, 7, 14, 30, 90, 500, and 2500 days of age were incubated for various times in artificial CSF containing 3 nM-[³H]thymidine at 37°C under 95% O₂-5% CO₂. When slices from all brain regions of 2-day-old rabbits were incubated in [³H]thymidine for 30 min, tissue-to-medium ratios of ³H were between 2 and 4 and declined with age, and the percentages of the total ³H in perchloric acid homogenates of brain slices as [³H]DNA were 26–29%, declining to low levels with age. However, at all ages and in all regions studied, 41 -88% of the ³H within the slices was phosphorylated. After homogenization and subcellular fractionation of the brain slices incubated in [³H]thymidine for 30 min, the highest percentage of [³H]thymidine phosphates plus [³H]DNA was present in the nuclear (crude and purified) and mitochondrial fractions of all brain regions. The [³H]DNA content in the nuclear and mitochondrial fractions declined with age, but the percentage of [³H]thymidine phosphates did not. Thymidine phosphates were synthesized from thymidine in all brain regions tested throughout the entire life span.     

Metabolism of Deoxyuridine in Rabbit Brain

Abstract: The metabolism of [³H]deoxyuridine by rabbit brain was investigated in vitro and in vivo . In vitro , brain slices from various regions of brain and from all age groups accumulated [³H]deoxyuridine from artificial CSF. Within the slices, a portion of the accumulated [³H]deoxyuridine was metabolized to [³H]deoxyuridine phosphate, with subsequent conversion to [³H]thymidine phosphate, and ultimately [³H]DNA. The percentage of the [³H]deoxyuridine phosphorylated and subsequently converted into [³H]DNA was highest at birth and declined to adult levels in 3-month-old rabbits. Thymidine, when added to the incubation medium with the [³H]deoxyuridine, was approximately 10 times as potent as unlabeled deoxyuridine in inhibiting the intracellular phosphorylation and conversion of [³H]deoxyuridine to [³H]thymidine phosphate in brain slices. In vivo , 2.5 h after intraventricular injection of [³H]deoxyuridine, over 90% of the [³H]deoxyuridine was cleared from the central nervous system at all ages. However, in both newborn and 3-month-old rabbits, approximately 40 and 12%, respectively, of the ³H remaining in brain was phosphorylated and converted to [³H]thymidine phosphates; and 11 and 4%, respectively, of the ³H remaining in brain was converted to [³H]DNA. These results show that both immature and mature rabbit brain is able to incorporate deoxyuridine into DNA. Thus, all the enzymes involved in this conversion, including thymidylate synthetase (EC 2.1.1.45), are present and active in brain throughout life.     

Thymidine incorporation into lake water bacterioplankton and pure cultures of chemolithotrophic (CO, H2) and methanotrophic bacteria

Abstract Incorporation of [³H]methyl thymidine into bacterial DNA was measured using samples of bacterioplankton from Lake Constance and pure cultures of CO, H₂ and CH₄-oxidizing bacteria. Thymidine was incorporated by Pseudomonas carboxydovorans, Paracoccus denitrificans, Methylosinus trichosporium, Methylomonas agile , and by various chemolithotropic or methylotrophic isolates from Lake Constance. Thymidine incorporation by bacterial cultures was stimulated by increasing concentrations of CO or H₂. Increased CH₄ concentrations stimulated thymidine incorporation by Ms. trichosporium only if the cells had been starved. In contrast to bacterial cultures, thymidine incorporation by bacterioplankton samples was not stimulated by increasing     

The proliferative response of rat liver parenchymal cells after partial hepatectomy A methodological study comparing flow cytometry of nuclear DNA content and in vivo and in vitro uptake of thymidine

Abstract. DNA synthesis in rat hepatocytes, from livers regenerating after 70% hepatectomy, was assessed by flow cytometric determination of nuclear DNA content and by incorporation of [³H]thymidine. Parenchymal liver cells were isolated by collagenase perfusion and low-speed centrifugation. Nuclei from the isolated cells were prepared for flow cytometry by a treatment with detergent, pepsin and RNase, and stained with ethidium bromide. Parallel samples of cells were incubated with [³H]thymidine and analysed for rate of incorporation of radioactivity into DNA and for labelling index determination.
The flow cytometric measure of the replicative response, i.e. the presence of cells with S-phase DNA content within the diploid and tetraploid cell populations, was compared with the incorporation of [³H]thymidine. For each of fourteen animals, including two control rats and twelve partially hepatectomized animals killed either before (at 13 hr after hepatectomy), at the onset (16 and 18 hr) or at the peak (24 hr) of regenerating activity, a fairly good correlation was found between the different methods. Satisfactory resolution of the flow cytometric detection of S-phase cells was indicated by a sorting experiment using an Ortho (system 50-H) cell sorter which demonstrated that after [³H]thymidine injection in vivo 88% of the diploid and 84% of the tetraploid S-phase nuclei were labelled, while labelling in the G¹-fractions was only 2 and 7%, respectively.     

Action of cortisol on the proliferation of rainbow trout fibroblasts

Abstract The effect of cortisol on the proliferation of the rainbow trout fibroblast cell line, RTG-2, was examined in synchronous and asynchronous cultures. When the transition from G₁ to S was synchronized by restoring serum to serum-deprived cultures, the addition of cortisol at the time of serum restoration delayed the entry of cells into S phase. However, if cortisol was added 24 h after serum restoration, at the G₁/S transition point, the subsequent peak of DNA synthesis was unaffected. In asynchronous cultures cortisol inhibited [³H]-thymidine and [³H]-uridine but not [³H]-leucine incorporation into acid-insoluble material. If the exogenous nucleoside concentration was raised, [³H]-thymidine but not [³H]-uridine incorporation continued to be inhibited by cortisol. This suggested that cortisol's effect on [³H]-thymidine incorporation reflected a change in entry into S phase and not just on thymidine uptake and metabolism. Cortisol inhibited the proliferation of RTG-2 in asynchronous cultures. At 1000 ng/ml of cortisol a reduction in cell number became apparent before the RTG-2 cultures were confluent, whereas at 100 ng/ml the reduction only became evident in confluent cultures. The synthetic antiglucocorticoid, RU 486, which acts at the level of the corticosteroid receptor, blocked the growth inhibition by cortisol. These results suggest that cortisol regulates rainbow trout fibroblast proliferation via the corticosteroid receptor and that the G₁/S transition is one point at which this regulation occurs.     

Local Re-Utilization of Thymidine In Normal Mouse Tissues As Measured With Iododeoxyuridine

Abstract. Thymidine (TdR) and its analogue, iododeoxyuridine (IdUdr), were used to quantitate nucleoside re-utilisation in vivo. Significantly different results are obtained, however, depending upon what form of isotopically labelled iododeoxyuridine is used. No measurable local thymidine re-utilization was found in mouse thymus, spleen or bone marrow when the retention of [³H]IdUdR was compared with [¹⁴C]TdR. On the other hand, significant differences were found between the retention of [¹²⁵I]IdUdR and [³H]IdUdR, which is attributed to de-iodination of iododeoxyuridine. Some thymidine re-utilization was found in duodenum using both [³H]IdUdR and [¹²⁵I]IdUdR. Information on the in vivo distribution of TdR and the contention that a large degree of TdR re-utilization in the thymus is evidence of extensive cell death must be re-interpreted in the light of these results. In addition, evidence for little or no local re-utilization in some tissues will greatly simplify the use of [¹¹C]TdR as an imaging agent for measuring tissue proliferation in vivo with positron emission tomography (PET).     

A Novel Photoaffinity Ligand for Kainate Sites Labels Two Polypeptides with Molecular Mass 45 and 33.5 kDa in Chick Cerebellum

Abstract: The synthesis of (2 S ,3 S ,4 S )-4-[1-(4-azidobenzamidomethyl)ethenyl]-2-carboxy-3-pyrrolidineacetic acid (ABCPA) is described. This novel kainic acid analogue, bearing a photolabile functionality on the isopropenyl side chain, was proven to be a good inhibitor of [³H]CNQX and [³H]kainic acid binding on chick cerebellar membranes. [³H]ABCPA was photoaffinity cross-linked on the membrane fraction of chick cerebellum. Electrophoretic analysis with sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed two major radioactive bands with apparent molecular masses of 45 and 33.5 kDa. [³H]ABCPA incorporation in both bands was completely blocked by 2 m M CNQX. When photoaffinity labeling was performed in the presence of 2 m M kainic acid, incorporation of [³H]ABCPA was blocked by ∼70% in the 45-kDa band and by 18% in the 33.5-kDa band. Incorporation of radioactivity in both bands was blocked by ∼30% with 10 m M glutamate.     

[3H]thymidine labels less than half of the DNA-synthesizing cells in the mouse tumour, carcinoma NT

Abstract. We have studied carcinoma NT, a transplantable mouse adenocarcinoma of spontaneous origin. Cells labelled with [³H]thymidine ([³H]TdR) were restricted to a narrow zone around the periphery of this tumour and were also found in rings up to 50 μ m wide, around isolated blood vessels in the central necrotic area. Labelling with [³H]deoxyuridine ([³H]UdR), another DNA synthesis precursor, produced a very different pattern. The labelled zone around the periphery was much wider than with [³H]TdR, and [³H]UdR labelled cells were found up to 110 μ m from isolated vessels. [³H]iododeoxyuridine ([³H]IUdR) gave the same pattern of labelling as [³H]UdR. In the heavily labelled zone, within 1 mm of the tumour periphery, the labelling index (LI) was 51% after [³H]UdR or [³H]IUdR injection, and only 36% with [³H]TdR.
The data show that at least half of the DNA-synthesizing cells in this tumour did not incorporate [³H]TdR. Previous workers reported cell loss factors for carcinoma NT of 60% calculated from [³H]TdR labelling data and 30% from the rate of loss of [¹²⁵I]UdR. The present work suggests that calculations based on [¹²⁵I]UdR data are more likely to be accurate for carcinoma NT than those using [³H]TdR data.     

Cell Position Dependence of Labelling Thymidine Nucleotides Using the De Novo and Salvage Pathways In the Crypt of Small Intestine

Abstract. About twice as much tritiated thymidine ([³H]TdR) is taken up by cells at the bottom of the crypt of the small intestine as by the rapidly cycling mid-crypt cells. However, the uptake of tritiated deoxyuridine ([³H]UdR) is even throughout the crypt.
Exogenous thymidine is incorporated about four times and eight times more efficiently than deoxyuridine by the cells in the mid-crypt and cells at the bottom of the crypt, respectively. However all S phase cells in the crypt appear to be capable of using either precursors, i.e. either the de novo or salvage pathway.
Since methotrexate (1 or 5 mg/kg) inhibits (at 5 mg/kg completely) the uptake of [³H]UdR, but has no effect on [³H]TdR uptake, the de novo and salvage pathways appear to be independent. Within the precision of the methods used in the experiments the 3 hr inhibition of the de novo pathway of deoxythymidylic acid (dTMP) synthesis by methotrexate does not produce any increase in utilization of the salvage pathway measured by incorporation of [³H]TdR into DNA. the increased efficiency of thymidine utilization by crypt base cells is not attributable to (i) differences in accessibility of thymidine; (ii) differences in the rate of DNA synthesis or (iii) the size of the nuclei.     

Double labelling of tissue combining tritiated thymidine autoradiography with immunodetection of bromodeoxyuridine: the autoradiographic significance of inhibition of thymidine incorporation into DNA by bromodeoxyuridine given simultaneously

Abstract We describe a reproducible method for combining tritiated thymidine ([³H]TdR) autoradiography with immunoperoxidase detection of bromodeoxyuridine (BrdU) in paraffin-embedded tissues. The technique has been used to examine, in mouse tongue epithelium, the inhibition of incorporation into DNA of [³H]TdR by a simultaneous injection of BrdU in the doses that both compounds are likely to be used in cell proliferation studies. The significance that this inhibition has on prolongation of autoradiograph exposure times, to ensure that all cells that incorporate [³H]TdR are scored as positive, in particular the most lightly labelled cells, has been quantified.
The inhibition of uptake into DNA of [³H]TdR from 0.23 to 1.85 MBq (6.25 to 50 μCi) per animal, produced by a simultaneous injection of 2.5 mg BrdU shows a linear, dose-dependent relationship. Provided the injected dose (in μCi per animal) multiplied by the autoradiographic exposure time (in days) is greater than a value of 700, then all cells that are labelled after incorporation of [³H]TdR alone are also labelled after simultaneous double labelling, despite the latter producing a lower average grain count.     

Proliferation of cells undergoing chondrogenesis in vitro

Abstract. Continuous exposure of chicken embryo limb bud mesenchyme cells undergoing chondrogenesis in vitro to [³H] thymidine ([³H]TdR) revealed that more than 90% of the cells synthesized DNA at least once during 120 h of culture. When cells were exposed to [³H]TdR for 24 h beginning at various times throughout the culture period, the percentage of cells which incorporated [³H]TdR during each period was approximately 92%. However, when the period for incorporation of radioisotope was limited to two hours, the number of cells which incorporated [³H]TdR was found to decline during chondrogenesis in vitro. This decline was coincident with the appearance of extracellular matrix material and occurred in those cells which had, and had not, expressed the cartilage phenotype.
We conclude from these studies that (1) practically all of the cells continue to proliferate while chondrogenesis is occurring in vitro, (2) there is an increase in the length of the cell cycle during chondrogenesis in vitro, and (3) withdrawal from the cell cycle is not required for differentiation of mesenchyme into cartilage.     

DNA POLYMERASE ACTIVITY FROM THE BRAIN OF OCTOPUS VULGARIS LAM

Abstract— The DNA polymerase of the optic lobes of Octopus vulgaris Lam has been solubilized and some of its properties have been investigated. The enzyme catalyses the incorporation of [³H]thymidine triphosphate into a DNase-sensitive, acid-insoluble product in the presence of DNA, the four deoxynucleoside triphosphates and Mg²⁺. The pH optimum is between pH 9.2 and 9.6, the optimal concentration of Mg²⁺ is between 10 and 30 mM and the optimal temperature is in the range 27°-30°C. The content of enzyme per mg protein is highest in the vertical and optic lobes and lowest in the suboesophageal lobe.     

Differentiation-Dependent Decline of DNA Synthetic Activities in Naegleria gruberi

SYNOPSIS. DNA of Naegleria gruberi strain NEG, grown in axenic culture, forms a band at a density of 1.6912 in CsCl gradient and has a GC content of 31.8%. Incorporation of [³H]thymidine into DNA is much reduced in differentiating Naegleria immediately after the stimulation to transforms, primarily because of the reduction in thymidine uptake by differentiating cells. In addition, there is a marked decrease in the rate of incorporation of [³H]thymidine and [³H]uracil into DNA at from 45 to 60 min after the stimulation for differentiation. This decrease in the rate of precursor incorporation into DNA appears to be due to the differentiation-dependent cessation of nuclear DNA synthesis. The differentiated phenotype (the flagellate) emerges at ∼ 70 min after the stimulation, and over 90% of the population differentiates within the next 30 min. Synthesis of mitochondrial DNA is detectable until 190 min after the stimulation. Since the S phase of Naegleria lasts ∼ 180 min, some cells in the population must cease synthesizing nuclear DNA in the middle of the S phase.     

Localization and Mechanism of Thymidine Transport in the Central Nervous System

Abstract: The localization and mechanism of thymidine and deoxyuridine transport in the central nervous system were studied in vivo and in vitro . Previous studies have shown that thymidine enters brain from blood in part via the CSF. In vitro , isolated adult bovine cerebral microvessels, which readily concentrated and phosphorylated deoxyglucose, were unable to concentrate thymidine and deoxyuridine. In vivo , [³H]thymidine (0.2 μ M ) and [³H]deoxyuridine(0.4 μ M ) were not extracted more readily than [¹⁴C]sucrose in a single pass through the cerebral circulation of rats. In vivo , [³H]thyrnidine retention in CSF and brain after entry from blood was increased when the efflux of [³H]thymidine from CSF and the phosphorylation of [³H]thymidine in brain were depressed by the intraventricular injection of unlabeled thymidine. These studies and previous work suggest that the transfer of thymidine (and deoxyuridine) through the blood-brain barrier in either direction must be extremely low. The present studies are consistent with the postulate that thymidine is transported by an active transport system in the choroid plexus that transfers thymidine from blood into the CSF; from the CSF, the thymidine enters brain cells and is phosphorylated.     

Acinar cell proliferation in the parotid and submandibular salivary glands of the neonatal rat

Abstract. Although the rat salivary glands are deficient in acini at birth, acinar cells proliferate rapidly during the early post-natal period. The pattern of acinar cell proliferation was analysed in the parotid and submandibular glands of neonatal rats from day of birth until day 34. Mitotic and [³H]thymidine ([³H]TdR) labelling indices of the two glands show distinctly different patterns. Analysis of cell division in the rat parotid gland demonstrated a peak of mitotic index at 14 days (2.9 ± 0.4%) and labelling index at 16 days (25.2 ± 2.1%). Submandibular gland acinar cell proliferation reaches a zenith between 7–8 days; labelling index (14.2 ± 1.1%) and mitotic index (2.3 ± 0.3%). Cell proliferation decreases rapidly in both glands after reaching a peak in activity. Gland size increases more rapidly in the submandibular gland which correlates with the earlier shift from cell proliferation to differentiation which occurs in this organ. Circadian rhythms of [³H]TdR incorporation were also investigated in this study. A circadian rhythm of [³H]TdR incorporation into DNA occurs at 15 days after birth with a peak at 06.00 hours in both glands and a trough occurring at 15.00 hours in parotid gland and 18.00 hours in the submandibular gland. Determination of specific activity of DNA (ct/min per μg DNA) on days 8, 10, 12, 13, 14, 15, and 16 after birth at 06.00 and 15.00 hours indicated that a circadian rhythm in [³H]TdR incorporation into DNA began on day 14. The developmental switch from suckling to solid food may be an initiating factor in the sychronization of the circadian rhythm in cell proliferation.     

Uptake of [3H]Dopamine into Dopaminergic and Noradrenergic Neurones of the Isolated Neurointermediate Lobe of the Rat Hypophysis. Effects of Desipramine and Nomifensine

Abstract: The isolated neurointermediate lobe (NIL) of the rat hypophysis accumulates [³H]dopamine from the incubation medium. Column chromatographic analysis showed that 92% of the tissue radioactivity was contained in the catecholamine fraction. [³H]Dopamine represented 70% and [³H]noradrenaline 30% of the [³H]catecholamines. Desipramine (1 μM) prevented the formation of [³H]noradrenaline without affecting the storage of [³H]dopamine. Nomifensine (10 μM) blocked the storage of [³H]dopamine and [³H]noradrenaline. Thus, in the NIL, [³H]dopamine is taken up into dopaminergic and noradrenergic neurones. In the latter, [³H]dopamine is converted to [³H]noradrenaline, indicating a significant dopamine β-hydroxylase activity in the NIL tissue. A selective labeling of the dopamine stores with [³H]dopamine can be achieved in the presence of desipramine.     

DNA-synthesizing cells in oral epithelium have a range of cell cycle durations: evidence from double-labelling studies using tritiated thymidine and bromodeoxyuridine

Abstract Mouse tongue epithelium is characterized by a circadian variation in the number of DNA-synthesizing cells (labelling index, LI). Cells undergoing DNA synthesis were labelled with tritiated thymidine ([³H]TdR) at 0300 (peak LI) or 1200 h (low LI). The fate of these cells was assessed by injecting animals with bromodeoxyuridine (BrdU) at intervals from 12–48 h after [³H]TdR, to follow them from one cell cycle to the next. Labelling was revealed by combining [³H]TdR autoradiography with immunoperoxidase detection of BrdU in the same sections.
A single peak in the appearance of double-labelled cells was seen at 44 h, if [³H]TdR was given at 1200 h; following [3H]TdR at 0300 h, a peak of double labelling was seen at 48 h with the possibility of smaller peaks at 24 h and 36 h.
These results show that the 24 h periodicity in LI in this tissue is associated with a predominant cell cycle duration of 44–48 h, but that a few cells cycle more quickly. Double labelling with [³H]TdR and BrdU provides a useful method for establishing cell cycle duration by labelling S-phase cells in successive cell cycles.     

Changes of Nucleic Acids and Proteins in the Oocysts of Eimeria tenella during Sporulation

SYNOPSIS. The total content of DNA in Eimeria tenella , estimated at 5.8 × 10⁻¹² gm/oocyst, varies little during sporulation. Its buoyant density is 1.682 gm/cm³, reflecting a G + C content of ∼41%. Thymidine is not incorporated into any TCA insoluble fraction of sporulating oocysts, but radioactivity from [³H]uridine and [³H]deoxyuridine are incorporated into RNA at a linear rate during the first 5 hr of sporulation. The labeled RNA, found mainly in the paranuclear bodies of newly formed sporozoites, contains ∼0.15 nmole [³H]uridine/10⁶ oocysts at the completion of sporulation. One nmole of leucine is incorporated into the hot TCA insoluble fraction of 10⁶ oocysts during the first 7 hr of sporulation after an initial lag. The incorporated amino acid is mainly in the cytoplasm of the sporozoites, and an analysis by SDS-gel electrophoresis reveals most of the radioactivity in a narrow band with a molecular weight of ∼50,000 daltons. Incorporation of uridine and leucine, however, can be totally suppressed by respiratory inhibition. Further analysis of the proteins in the oocysts reveals that the total protein content remains relatively unchanged at 2.64 × 10⁻¹⁶ gm/oocyst during sporulation, but there is a shift of 13–14% of total protein from the soluble cytoplasm to the 15,000 g pellets. By polyacrylamide gel electrophoresis, a major protein band. possibly a glycoprotein, is shown in the soluble cytoplasm of unsporulated oocysts. This band disappears during sporulation.     

CAPABILITY OF TUBULIN AND MICROTUBULES TO INCORPORATE AND TO RELEASE TYROSINE AND PHENYLALANINE AND THE EFFECT OF THE INCORPORATION OF THESE AMINO ACIDS ON TUBULIN ASSEMBLY

Abstract— Incorporation of [¹⁴C]tyrosine into the C-terminal position of α-tubulin of rat brain cytosol was 10-fold higher for non-assembled than for assembled tubulin. The incorporation into tubulin from disassembled microtubules was higher than into non-assembled tubulin; therefore, the low incorporation into microtubules was not due to a lower acceptor capacity of their tubulin constituent.
[¹⁴C]Tyrosine was released from assembled and non-assembled [¹⁴C]tyrosinated tubulin by the action of an endogenous carboxypeptidase. Release from non-assembled tubulin was shown by incubating a tubulinyl-[¹⁴C]tyrosine preparation in the presence of CaCl₂ at a concentration that abolished microtubule formation. Release from microtubules was inferred from the observation that the percentages of [¹⁴C]tyrosine released and the decrease of the specific radioactivity of the recovered microtubules were practically identical and did not change after a 10-fold dilution of the incubated microtubules.
[³H]Phenylalanine was released from a preparation of tubulinyl-[³H]phenylalanine also by an enzymatic activity.
The capacity of a tubulin preparation to incorporate tyrosine was increased 43% by pre-treatment with endogenous carboxypeptidase.
Tubulin tyrosinated in vitro was assembled to the same extent as native tubulin. After a mixture of tubulinyl-[¹⁴C]tyrosine and tubulinyl-[³H]phenylalanine was partially assembled, the ratio of ¹⁴C/³H found in the microtubules was the same as in the non-assembled tubulin fraction.