`
`Experimental Cell Research 45, 1584 6'6 (1.966)
`
`© 1967 by Academic Press Inc.
`
`COMPARATIVE EFFECTIVENESS OF INTRAPERITONEAL
`
`AND INTRAMUSCULAR 3H-TDR INJECTION
`
`ROUTES IN MICE1
`
`M. R. SKOUGAARD2 and P. A. STEWART3
`
`Biology Department, Brookhaven National Laboratory, Upton, N. Y. 11973, U.S.A.
`
`Received August 8, 1966
`
`IN tritiated thymidine (3H-TDR) autoradiography, intraperitoneal injection
`of label is very frequently used, especially in small animals. The kinetics
`of the numerous processes involved between injection and incorporation into
`nuclear DNA are 11ot well established, however. Recent work on the marmo-
`
`set [14] has shown that only a small fraction of the injected “H-TDR is incor-
`porated into DNA, and that the yield of autoradiographic grain counts
`depends on the time course of tracer concentration in the bloodstream.
`3H-TDR is a major part of the nonvolatile 3H—plasma activity only during
`the first few minutes after injection. Significantly different results were ob-
`tained when intramuscular (IM) rather than intraperitoneal (IP) injection
`was used The four-factor model introduced by Quastler [9] predicts such
`results, but has been tested in detail only for IP injection in the mouse [17].
`The kinetics of thymidine incorporation into DNA after IP injection into
`mice have also been studied indirectly by a number of workers. On the
`
`basis of time taken for maximum crypt cell labeling to occur, Quastler and
`Sherman [11] concluded that IP injected 3H-TDR was completely absorbed
`from the peritoneal cavity by 16 min post—injection. Quastler and Kember
`[10] measured the tritium content of peritoneal washings as a function of
`
`time after injection, and concluded that uptake of 3H—TDR into the circula-
`tion followed a double exponential curve with half-times of approximately
`
`g— min and 5 min. They also found that maximum autoradiographic labeling
`had occurred in the intestinal crypt cells 15 min after the injection. Staroscik
`et al.
`[15] found uptake of 3H—TDR still occurring into mouse mammary
`
`1 Research carried out at Brookhaven National Laboratory under the auspices of the US.
`Atomic Energy Commission.
`2 Present address: Department of Periodontology, Royal Dental College, Copenhagen, Den-
`mark.
`3 Present address: Division of Medical Science, Brown University, Providence, R.I. 02912,
`U.S.A.
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 1
`Mylan Pharms. Inc. V. Astrazeneca AB lPR20l6-01324
`
`
`
`Intraperitoneal and intramuscular 3H-TDR injection routes in mice
`
`159
`
`gland tumors 40 min after injection. They estimated the rate of uptake by
`evaluating the derivative of the curve of autoradiographic labeling versus
`time after injection and found a rapid initial phase followed by an exponen-
`tial curve with a half—time of about 10 min. Bresciani [2] used the derivative
`
`of the autoradiographic labeling curve from the same tissue to estimate the
`activity of 3H—TDR in the bloodstream. He found an approximately exponen-
`tial decrease with a half-time of about 7 min. Rubini et al. [13] carried out
`
`studies of 3H~activities measured directly in the blood plasma of man follow-
`
`ing intravenous injection of 3H-TDR. They found a double exponential
`
`disappearance curve with half—times of 0.2 and 6 min.
`In the present study, intraperitoneal and intramuscular injection routes
`for 3H-TDR in mice have been compared with respect to the following para-
`meters:
`
`(1) Time course of blood plasma activity level of total tritium, volatile
`
`tritium, nonvolatile tritium and tritiated thymidine.
`(2) Final levels of labeling intensity (mean grains per labeled nucleus) in
`oral epithelial and intestinal crypt cell autoradiographs.
`
`MATERIALS AND METHODS
`
`Ninety-six Walter Reed Hospital Swiss pathogen—free mice from the laboratory
`colony were used for the experiment [5]. Three— to 4—month—old males from 26 litters,
`ranging from 27 to 36 g were selected. Tritiated thymidine with a specific activity
`of 6.5 c/mole was heated to 36°C and injected at a dose of 2 no per g body Weight,
`either IP or IM, into the right thigh muscle mass.
`Blood samples were taken at intervals of 1 to 90 min after injection, only one sample
`being taken from each animal. One-minute samples were obtained by decapitating
`the animals over a heparinized centrifuge tube. Samples of 0.5 ml were obtained
`in 3 to 5 sec in this way, but the blood tended to coagulate extremely rapidly. There-
`fore this technique was only used for the 1—min samples, for which the time factor
`was critical. The later blood samples were taken by aspiration through a thin hepa-
`rinized glass pipette inserted behind the eyeball into the infraorbital venous plexus.
`Samples of 0.5 ml were obtained within 20 to 30 see by this technique. All samples
`were immediately sealed, chilled to 4°C and centrifuged 30 min at 3450 x g. Expo-
`sure of samples to air was kept at a minimum to avoid loss of volatile tritium-con-
`taining components.
`Aliquots of 100 pl of plasma were transferred to counting vials and 3.9 ml ethyl
`alcohol plus 16 ml of 0.3 per cent PB D—xylene added for measurement of total plasma
`tritium activity. Volatile tritium activity was determined on another 100 pl aliquots
`by distilling the volatile components into a counting vial chilled with dry ice and
`acetone (Fig. 1). The sample was placed in the small lower vial, which was imme-
`diately clamped to the closely-fitting upper vial. The lower vial was then heated in
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 2
`
`
`
`160
`
`M. R. Slcougaard and P. A. Stewart
`
`a boiling Water bath while the upper vial was surrounded by the dry ice—acetone freez-
`ing mixture. After 5 min the whole unit was disassembled and the upper vial sealed
`immediately after addition of ethyl alcohol and PB D-xylene.
`Test runs with known amounts of tritiated water showed 100 per cent recovery
`by this technique. All samples were counted in a liquid scintillation spectrometer
`for a period of time sufficient to reach a minimum of 104 counts. Background and
`standard counts were determined for each day of counting, and results corrected
`accordingly. The efficiency of the scintillation spectrometer was found to be 8 per
`cent. Nonvolatile tritium activity of plasma was calculated by subtracting volatile
`activity from total activity as determined from these two aliquots of each sample.
`For determinations of the per cent of 3H-TDR in the nonvolatile tritium activity,
`samples of pooled blood from four animals at each specific time were centrifuged.
`After protein precipitation with trichloracetic acid, plasma samples were centrifuged
`at 3450 X g for 30 min. The supernatant was neutralized to pH 7.0 with 1.0 N NaOH,
`lyophylized and reconstituted to 200 Ml with distilled water. A 10 ,ul aliquot of this
`solution was then analyzed by ascending paper chromatography using the upper
`layer
`from an ethyl acetate:water:formic acid:60:35:5 mixture [4]. Additional
`thymidine carrier was added to the sample spot before chromatography. The thy-
`midine spot was identified under ultraviolet light, cut out of the paper, eluted into
`Bray’s solution [1] and counted in the liquid scintillation spectrometer.
`The activity of 10 ul of the reconstituted sample was also determined by letting
`it dry into a spot on filter paper, cutting out this spot, and treating it exactly as
`the 3H—TDR spot was treated. The percentage of 3H—TDR in the total nonvola-
`tile plasma tritium activity was then calculated from the ratio of the two counts.
`For autoradiography, small pieces of intestine and sublingual tissue were fixed
`in Carnoy’s solution immediately after sacrifice, stained by conventional Feulgen
`technique and dissected to provide squash preparations of the epithelia on micro-
`scope slides. Kodak NTB liquid emulsion was used, and emulsion—covered slides were
`exposed for appropriate periods of time in the cold, with dessicant [3]. After photo-
`graphic processing, grains over the nuclei were counted and the resulting data pro-
`cessed by an IBM 7094 computer.‘
`
`RESULTS AND DISCUSSION
`
`Time course of tritiated components in bloodstream
`
`The total volatile and nonvolatile (by subtraction) plasma activities are
`plotted against time in Figs. 2—4. Each point on these curves represent the
`average value for the 4 mice sacrificed at the indicated time after injection.
`In the IP injected mice, total 3H-activity reached its maximum value of
`2 X104 cpm/100 ,ul plasma almost
`immediately following injection, and
`remained at this level during the rest of the experiment. Volatile 3H-activity
`
`1 The program for the computer was devised by Mr K. Thompson of the Biology Department,
`Brookhaven National Laboratory.
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 3
`
`
`
`Intraperitoneal and intramuscular 3H— TDR injection routes in mice
`
`161
`
`increased at a slower rate, but eventually reached a plateau level indis—
`tinguishable from that of the total activity. The nonvolatile activity, measured
`as the difference between total and volatile, increased to a maximum and
`
`then fell off slowly to negligible values. The form of the curves for the IM
`
`cpm/I00atPLASMA6”0,-————-——-E-,-—————-—-—_._
`TRITIUMACTIVITY
`
`
`io
`
`40
`20
`TIME AFTER INJECTION (MINUTES)
`
`50
`
`Fig. 2.
`
`Fig. 1.——Device used to collect volatile plasma components in the liquid scintillation counting
`vial (B). When sealed by the clamp the two vials, A and B, fit closely. a, dry ice acetone; b, vial;
`c, boiling water.
`
`Fig. 2.—Semilogarithmic graph demonstrating the total and the volatile 3H—activities in the blood
`plasma vs. time after IP injection of 3H-TDR. Each point represents the average activity for
`the 4 mice sacrificed at the time interval in question. 0-0, total plasma “H-activity; O—o,
`volatile plasma "H—activity.
`
`injected mice was similar, but all changes were slower. It was possible in
`this case to observe the increase in total plasma activity during the first
`5 min after
`IM injection, and the rate of disappearance of nonvolatile
`activity was much less than after IP injection. It is evident from the figures
`that the decrease in nonvolatile 3H—activity was exponential following IP as
`
`well as IM injection and dropped to negligible values by 30 min after IP or
`60 min after IM injection.
`Attempts were made to determine total plasma activity immediately after
`IP injection. In all cases results on the order of 2 X104 cpm/100 /J]. plasma
`were obtained, so that we may take this number as a measure of the initial
`
`1 1 — 67 1 8 1 1
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 4
`
`
`
`162
`
`M. R. Skougaard and P. A, Stewart
`
`total plasma activity. If all the injected 3H—TDR (2 ‘uc/g) were distributed
`uniformly over the extracellular Water phase, which we take to be 20 per
`cent of the body weight [6, 16], the expected initial activity would be 2 X105
`cpm/100 ,ul plasma. This suggests that approximately 90 per cent of the IP
`
`A I
`
`5
`
`
`
`
`
`TRITIUMACTWITYcpm/I00plPLASMA 5Of
`
`7
`
`I
`
`I02
`
`Q.
`
`PLASMA 5e.-
`TRITIUMACTIVITYcpm/I00,u.].
`
`
`
`
`I
`I
`40
`20
`TIME AFTER INJECTION (MINUTES)
`
`I
`
`I
`
`I
`60
`
`3
`I0 O
`
`I
`
`I
`
`I
`I
`40
`20
`TIME AFTER INJECTION (MINUTES)
`
`I
`
`I
`60
`
`Fig. 3.
`
`Fig. 4.
`
`Fig. 3.—Semi1ogarithmic graph demonstrating the total and the volatile 3H—activities in the blood
`plasma vs. time after IM injection of 3H-TDR. Each point represents the average activity for
`the 4 mice sacrificed at the time interval in question. 040, Total plasma 3H—activity; O40,
`volatile plasma 3H—activity.
`
`Fig. 4.—Semilogarithmie graph showing the nonvolatile 9H-activity in the plasma vs. time after
`IP and IM injections. O—O, IP injection; o—o, IM injection.
`
`injected 3H—TDR is catabolized, probably Via the hepatic portal system and
`the liver [8] and not released into the systemic circulation.
`Results of the paper-chromatographic analysis of the plasma tritium acti-
`vities are plotted in Fig. 5. The 3H-TDR activity constituted a rapidly decreas-
`ing percentage of the nonvolatile activity, and decreased more rapidly after
`IP than after IM injection. This difference might be expected to result from
`the more direct route to liver catabolic mechanisms from intraperitoneal sites
`than from intramuscular.
`
`Multiplying the 3H-TDR percentages from Fig. 5 by the nonvolatile acti-
`vities for corresponding times (Fig. 4) gives the actual plasma 3H-TDR values
`expressed as cpm/100 ,ul plasma. In Fig. 6 these 3H—TDR concentrations
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 5
`
`
`
`Intruperitoneal and intramuscular 3H-TDR injection routes in mice
`
`163
`
`are plotted as a function of time after injection. Following IP injection, the
`plasma 3H-TDR curve is a rapidly decreasing exponential, with a half-time
`of about 2.9 min. It can be closely approximated by a single exponential
`function of the form 2.0 ><104 ><e‘““5. After IM injection, plasma 3H-TDR
`
`I00
`
`so
`
`E
`E so
`E
`
`40
`
`2o--
`
`\
`
`
`
`0
`
`o
`
`_.\\~
`.‘.‘~.C'
`‘-—I~.
`20
`60
`4o
`TIME AFTER INJECTION (MINUTES)
`
`K\
`
`\
`
`\\
`4
`2‘
`I
`.,,1o —
`SO.
`fit
`
`:9\
`
`E:
`E
`E lo’
`,5
`<E
`
`E
`E
`F
`
`.02
`
`H3
`
`‘M
`
`\
`\
`
`\
`
`\
`
`\
`\
`I
`I
`L\__ I
`40
`20
`TIME AFTER INJECTION IMINUTESI
`
`I
`
`I
`60
`
`Fig. 5.
`
`Fig. 6.
`
`Fig. 5.—The 3H—TDR as per cent of the nonvolatile “H-activity vs. time after IP and IM injec-
`tions. O—O, IP injection; O—O, IM injection.
`
`Fig. 6.—3H~TDR plasma activity vs.
`O—O, IM injection.
`
`time after IP and IM injections. 0&0, IP injection;
`
`activity increased to a maximum of 9 X103 cpm/100 ,ul plasma at about 5
`min after injection, followed by an exponential decrease. This curve can be
`approximated by a double exponential of
`the form 2.5 ><104 ><(e'““'“—
`e-t/2.5).
`Since in both cases negligible plasma 3H-TDR activity was left after
`60 min, the integral of either function from zero to 60 min is a measure of
`the cumulative amount of 3H-TDR available for incorporation after a single
`injection in counts per 100 ,ul plasma. The same quantities can be calculated
`by measuring the areas under the curves of 3H—TDR activity. The results of
`these calculations show that the cumulative 3H-TDR available was 92 X103
`counts/100 pl plasma after IP and 120 X103 counts/100 ,ul after IM injection.
`In summary, these results show that after IP injection, plasma 3H-TDR
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 6
`
`
`
`164
`
`M. R. Skougaard and P. A. Stewart
`
`levels rise extremely rapidly and then fall off exponentially. After IM injec-
`tion, plasma 3H-TDR levels rise more slowly to a maximum after about
`5 min, and then fall off exponentially, but more slowly than after IP injec-
`tion. The cumulative amount of plasma 3H«TDR available for incorporation
`
`TABLE I. Labeling intensity expressed as the average of the mean grain counts
`from autoradiographs of crypt and tongue epithelium following IP and IM
`injections of 3H-TDR.
`
`Each average for the crypt cell data is compiled from 5 mice and for the tongue
`epithelium from 4 mice.
`
`Cell type
`
`Crypt
`Crypt
`Tongue
`Tongue
`
`Injection
`route
`
`Average mean
`grain count
`(grains/nucleus)
`
`Standard error
`(grains/nucleus)
`
`IP
`IM
`IP
`IM
`
`24.5
`13.7
`17.2
`15.5
`
`1.4
`1.5
`2.5
`2.7
`
`into nuclear DNA is approximately the same in each case, in spite of these
`marked differences in kinetics.
`
`Autoradiograph grain counts
`
`The grain count distributions from the autoradiographs obtained from
`
`10 mice sacrificed 1 hr after either IM or IP injection of 3H-TDR all showed
`
`the same general pattern. The numerous problems involved in interpreting
`precisely such a distribution have been discussed previously [7, 12, 14].
`For the present purposes we have arbitrarily taken the mean grains per
`nucleus for nuclei with 4 or more grains as the parameter to characterize
`labeling intensity. The results are listed in Table I. It will appear that
`there is no marked difference in labeling intensity between tongue epithelial
`cells in IP and IM injected mice, as would be expected from the figures ob-
`tained above for the amounts of 3H-TDR available. In the IM case, there is
`
`also no significant difference between tongue epithelial cells and intestinal
`crypt cells, as would be expected if both cell types are taking up 3H-TDR
`from the same plasma, and have similar DNA synthesis rates.
`There is, however, a significant difference between the labeling intensities
`of crypts cells and tongue epithelial cells in mice after IP injection, the crypt
`cells having roughly twice the labeling intensity of the tongue epithelial cells.
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 7
`
`
`
`Intraperitoneal and intramuscular 3H-TDR injection routes in mice
`
`165
`
`The simple fact of proximity of the crypt cells to the high concentration of
`3H-TDR injected may provide a partial explanation for this difference.
`
`Direct uptake of tracer from the peritoneal cavity into the tissue fluid sur-
`
`rounding the intestinal crypt cells might occur. More in line with the comments
`above concerning the importance of the hepatic portal system, however, is
`the notion that uptake from the peritoneal cavity into the intestinal circula-
`
`tion may be significant, both in channeling the 3H-TDR to the liver, for
`removal, and in supplying the intestinal crypt cells initially with a much
`
`higher effective dose than that eventually supplied to cells such as tongue
`
`epithelium more remote from the injection site.
`In all cases, the variability in the grain counts, expressed as the standard
`
`error of the mean grains per nucleus, is about 20 per cent. The reasons for
`such a high variation are still not clear [17]. There seems to be no evidence
`
`in this study that IM injection in mice results in a lower variability, in either
`
`cell type.
`
`SUMMARY AND CONCLUSIONS
`
`Blood samples were taken from 96 mice at various time intervals after
`intraperitoneal (IP) or intramuscular (IM) injections of 3H—TDR. The total
`volatile and nonvolatile 3H—plasma activities were determined in a liquid
`scintillation spectrometer. The 3H—TDR time curve was found by paper
`chromatography analysis of the nonvolatile plasma components. Following
`IP injections, the plasma 3H-TDR concentration decreased exponentially with
`a half—time of 2.9 min, and with effectively complete plasma clearance, by
`30 min. With IM injection, the maximum 3H—TDR concentration occurred
`after 5 min, followed by a slower exponential decrease than that after the
`IP injection. Following either route of administration, the cumulative 3H-
`TDR amount was approximately the same.
`Autoradiographic grain counts were carried out on squash preparations
`from tongue and crypt epithelium. The yield of grain counts was about the
`same in the tongue epithelium following IP and IM injection and in crypt
`cells following IM injection. In the crypt cells after IP injection, however,
`the mean grain count was about twice that in the other 3 groups. It is sug-
`gested that this difference might be due to the uptake of 3H—TDR from the
`peritoneal cavity into the intestinal circulation resulting in a higher effective
`dose for the gut compared to the tongue epithelium.
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 8
`
`
`
`166
`
`M. R. Skougaard and P. A. Stewart
`
`REFERENCES
`
`1. BRAY, G. A., Anal. Biochem. 1, 279 (1960).
`2. BRESCIANI, F., Intern. Congr. Biochem. Abstr. Papers, p. 45 (1964).
`3. ELIAS, J., Stain Technol. 39, 235 (1964).
`4. FINK, K., CLINE, R. E., HENDERSON, R. B. and FINK, R. M., J. Biol. Chem. 221, 425 (1956).
`5.
`INNES, J . R. M. and BORNER, G., Informal Report BNL 6744, Brookhaven Natl Laboratory,
`Upton, New York, 1962.
`6. NICHOLS, G., J}{., NICHOLS, N., WEIL, VV. B. and VVALLACE, W. M., J. Clin.
`1299 (1953).
`7. PELC, S. R., in L. F. LAMERTON and R. J. M. FRY (eds.), Cell Proliferation, p. 94. Blackwell
`Sci. Publ., Oxford, 1963.
`8. POTTER, V. R., in F. STOHLMAN (ed.), The Kinetics of Cellular Proliferation, p. 104. Grune
`and Stratton, New York, 1959.
`9. QUASTLER, H., in M. HAISSINSKY (ed.), Chemical and Biological Effects of Radiation, p. 149.
`Masson et Cie, Paris, 1963.
`10. QUASTLER, H. and KEMBER, N. F., Unpublished manuscript (1963).
`11. QUASTLER, H. and SHERMAN, F. G., Exptl Cell Res. 17, 420 (1959).
`12. QUASTLER, H. and WIMBER, D. R, Unpublished manuscript (1963).
`13. RUBINI, J. R., CRONKITE, E., BOND, V. P. and FLIEDNER, T. M., J. Clin. Invest. 39, 909
`(1960).
`14. SKOUGAARD, M. R., Acta Odontol. Scand. 22, 693 (1964).
`15. STAROSCIK, R. N., JENKINS, W. H. and MENDELSOHN, M. L., Nature 202, 456 (1964).
`16. STEELE, R., WALL, J. S., DE B000, R. C. and ALTSZULER, N., Am. J. Physiol. 187, 15 (1956).
`17. STEWART, P. A., QUASTLER, H., SKOUGAARD, M. R., WIMBER, D. R., WOLFSBERG, M. F.,
`PERROTTA, C. A., FERBEL, B. and CARLOUGH, M., Radiation Res. 24, 521 (1965).
`
`Invest. 32,
`
`Experimental Cell Research 45
`
`Astrazeneca Ex. 2128 p. 9