Studies on the fate of lymphocytes III. The migration and metamorphosis of in situ labeled thymic lymphocytesкод для вставкиСкачать
Studies on the Fate of Lymphocytes 111. THE MIGRATION AND METAMORPHOSIS OF IN SITU LABELED THYMIC LYMPHOCYTES ' RAYMOND G . MURRAY AND PHYLLIS A. WOODS Department of Anatomy and Physiology, Indiana University, Bloomington, Indiana ABSTRACT The in situ thymus of the guinea pig was labeled locally with H3-thymidine and sections of various tissues analyzed by radioautography for the location of labeled cells which might have left the thymus from 30 minutes to three days after labeling. Labeled lymphocytes were located especially in the mesenteric lymph nodes and spleen, giving direct evidence for the migration of thymocytes under near-physiological conditions. Only a few cells were found in the bone marrow, and none in the liver. The identified cells were not seen to be undergoing subsequent proliferation, but about one-fourth of them were transforming to other cell types plasma cells, heterophil granulocytes, reticular cells, and macrophages. The fate of another portion of the cells seemed to be loss to the gut lumen through the mucosa. These data suggest that a normal function of the thymus of the adult guinea pig is the proliferation of a population of pluripotential stem cells which, after migration to other tissues, may further differentiate under appropriate stimulus. Possible applications of the local labeling technique are indicated. In the last few yeaxs there has been a heightened interest in the thymus with the work of Miller ('63) and others in establishing the thymus as a critical organ in the initiation of immunological competency. The part played by the thymus as a highly productive lymphatic tissue, acting as a contributor of cells, has also been stressed by many (Walker and Leblond, '58; Fichtelius, '60; Yoffey, '60; and many others). However, all the evidence in support of the thesis that the thymus acts as a supplier of cells for other tissues has been indirect from the following principal methods of study: ( 1 ) thyectomy of both newborn and adult animals, noting subsequent cellular changes (Gyllensten, '53; Beirring, '60; Metcalf, '60; Jankovic et al., '62), ( 2 ) cell population kinetics, studied by mitotic index (Kindred, '42; Leblond, '59; Sainte-Marie and Leblond, '64), ( 3 ) morphologic evidence of migration within the lymphatics and blood vessels of the thymus (SainteMarie and Leblond, '58, '64; Clark, '63; Weiss, '63) (4) transfusion of labeled cell suspensions from the thymus (Keohane and Metcalf, '58; Fichtelius, '60; Rankin, '60; Murray and Murray, '61; Diderholm, '61 ; Mims, '62), and (5) ontogenetic develANAT.REC., 150: 113-128. opment of the thymus and other lymphatic tissues (Miller, Marshall and White, '62; Papermaster et al., '64). In a recent review, Arnason et al. ('62) concluded, "The thymus appears to be a principle source of the small lymphocyte population of blood, spleen, and lymph nodes." Nevertheless, the critical demonstration that significant numbers of thymus cells actually migrate from the intact organ to other parts of the body has not been made (Yoffey, '62; Miller, '63). We propose to supply this demonstration, avoiding some of the objections to the transfusion experiments cited above, by providing more nearly physiological conditions. To do so we have: (1) left the thymus in situ, with its attachments to the circulation and lymphatics of the animal, and (2) injected directly, with minimum trauma, minute volumes of a label which is relatively innocuous and essentially permanent once acquired. The guinea pig was used because of the accessibility of the thymus. 1 This investigation was supported in part by Grant no. HE-06257 from the National Institutes of Health. N.S.F. Summer Graduate Teaching Fellow 1963 In partial fulfillment of the requirements for the dexree Master of Arts in Anatomy and Physiology, Indiana University. 2 113 114 RAYMOND G. MURRAY AND PHYLLIS A. WOODS Preliminary reports of this technique were given by Murray ('62) and Woods ('64). MATERIALS AND METHODS Twenty-seven guinea pigs from a colony maintained in OUT laboratory, with an average weight of about 400 gm, ranging from 293 to 598, were used for the study. For the injection of H3-thymidine (Schwarz Laboratories, specific activity, 6.0 C/mM, isotonic) into the left lobe of the thymus, the animals were anesthetized with Nembutal, and under sterile conditions a midline incision was made in the neck. Blunt dissection was used to carefully detach the lower part of the left lobe from the surrounding connective tissue, leaving the major vessels to the lobe intact. A ligature was placed around the vascular pole and was constricted just prior to the injection of the thymidine. Table 1 lists the animals, body weights, dose of thymidine administered, treatment of the circulation to the lobe during the injections, and intervals allowed before autopsy. The dose ranged from 4 yc to 83 yc, and was administered usually in about six small injections spread over a five minute interval in various portions of the lobe. The total volume injected was approximately 0.05 ml, and was injected with a 30 gauge needle in a microsyringe. Tyrode-moistened gauze was placed around the lobe during the injection and before loosening the ligature to absorb any H3-thymidine which may have come to the surface of the tissue from the injected point. Usually five minutes were allowed after the injection before the circulation to the lobe was restored, but sometimes the interval allowed was 15, and in seven animals the circulation was left open and no dissection around the lobe was performed for the injection of the label. After the injections the animals received Penicillin intraperitoneally, were sutured, and placed in recovery cages. The intervals between injection of thymidine and autopsy ranged from 30 minutes to five days (see table 1 for details). The animals were killed with ether. Duodenum, spleen, liver, kidney, Peyer's patch, mesenteric lymph node, lung, cervical lymph nodes from the right and left sides, both lobes of the thymus, and femoral bone marrow were examined. The procedures for radioautography are discussed in the previous paper (Murray and Murray, '64), as well as the conditions relating to the significance of grain counts. We have again used the value four grains or more to designate a positively labeled cell. OBSERVATIONS Labeling in the thymus Labeling efticiency. The injected lobe was adequately labeled (i.e. 20-40% of the cells labeled) when 25-50 I.IC were employed with brief ligation. Labeling was inadequate with 4 yc, and leakage was excessive without ligation, whether 50 or 83 LLCwere used. Although the local concentration of Ha-thymidine was probably higher than that following the customary systemic labeling procedures, the technique did not result in obvious damage to the cells. A comparison of the animals which received 4 yc with those receiving 83 yc showed no significant differences in cell death in the thymus. Patches of dead cells were occasionally noted, but these appeared to have resulted from handling rather than from irradiation damage. Although there are obvious drawbacks to occluding the circulation to the thymus, even for only five minutes, attempts to achieve a strictly local label were unsuccessful without this precaution. Care was taken that the occlusion was made only immediately preceding the injection and released immediately thereafter. Pattern of thymus label. The heaviest and most thorough label was found at the periphery of the cortex. At least 50% of the cells in the cortex were labeled with six or more grains two days after 50 yc of thymidine, and some of these cells were so densely labeled that their nuclei were completely obscured (see fig. 1 ) . Usually about 20% of the labeled cells found in the cortex had this very heavy label. Grain counts were not made in the thymus because of the large errors involved in the counting of these heavily-labeled cells, but it appeared that the average intensity of label was not strikingly reduced at three days over earlier intervals. Both lymphocytes and epithelial cells were labeled, as well as occasional devel- 115 FATE O F LOCALLY LABELED THYMOCYTES TABLE 1 Details of faCtOTS involved in treatment of the animals Dose of ThH3 Treatment of circulation to thymus during injections Interval before autopsy Animal number Weight of animal 4 30 min 30 min 1day 2 days 31 32 29 30 372 296 367 405 ligated for 5 min ligated for 5 min ligated for 5 min ligated for 5 min 25 1day 3 days 5 days 56 51 54 371 289 336 ligated for 15 min ligated for 15 min ligated for 15 min 41 1day 2 days 2 days 50 49 52 458 516 459 ligated for 15 min ligated for 15 min ligated for 15 min 50 30 min 1hr 1day 1day 2 days 2 days 3 days 3 days 3 days 3 days 38 47 34 44 33 35 36 37 48 53 406 264 501 267 345 395 298 511 238 251 ligated for 5 min free circulation ligated for 5 min free circulation ligated for 5 rnin ligated for 5 min ligated for 5 min ligated for 5 min free circulation ligated for 15 rnin 46 39 40 43 41 42 45 413 560 598 478 407 413 388 free circulation free circulation free circulation free circulation free circulation free circulation free circulation gm CC 83 Notes on special treatment 1hr 12 hr 12 hr 1day 2 days 2 days 3 days ~ ~~ both lobes injected partially thymectomized both lobes injected ~~ 1 All animals designated as “free circulation” showed thymidine leakage, and were discarded for the principal analysis of the fate of thymocytes. oping eosinophil myelocytes. Within the Hassall’s bodies, label was faund in immature eosinophil leukocytes and dead cells as well as in epithelial cells of the outer layers. The medulla was, in general, much less heavily labeled than the cortex, but with time, labeled small lymphocytes in the medulla increased, which would be consistent with the migration of cells from the cortex to medulla, as observed by Ernstrom (’63), and Sainte-Marie and Leblond (’64). Based on measurement of 100 labeled cells in each animal, a high inverse correlation between nuclear diameter and the length of time after injection of H3-thymidine is illustrated (figs. 2 and 3). The rise with time in the population of small lymphocytes was presumably by mitosis of the larger cells. This finding is in agreement with the evidence of Leblond and Sainte-Marie (’60), and Yoffey et al. (’61) that in the thymus the cellular proliferation is from the larger lymphocytes to small lymphocytes. Not only did the average size of labeled cells decrease, but the total per cent of labeled cells increased with time. Figure 4 illustrates the per cent of cells labeled in the thymus at the various intervals, and indicates an early labeling efficiency of about 3% at 30 minutes, which rises with time. By three days, in one instance, about half of the cells of the thymus were labeled. These percentages were calculated from counting in each animal about 500 cells selected from five different fields, including one area of dense cortex, one of medulla, and three intermediate fields. It was concluded from these measurements that a good labeling efficiency is attained by an injection of 25-50 I.IC of the 116 RAYMOND G. MURRAY AND PHYLLIS A. WOODS label, with the vasculature occluded for five minutes, and allowing an interval of two or three days before autopsy. Degree of confinement of label to the thymus The degree of label in the epithelial cells of the duodenal mucosa was chosen as a measure of the extent of systemic distribution of the H3-thymidine injected. This choice was based on the known ability of this tissue to acquire the label (Quastler and Sherman, ’59; and others) and on our observations in those cases where a ligature was not employed and some leakage occurred. These animals were readily distinguishable from those in which a ligature was used (figs. 5 and 6 ) . The duodenal epithelial cells were well labeled, and although label was also found in mesenteric lymph nodes, spleen and bone marrow, the grain counts over duodenal epithelial cells were as high as, and in most cases higher than, those seen elsewhere. Exceptions to this were occasionally noted in cervical lymph nodes and uninjected thymus lobe, but in both these sites there may have been some direct contamination with label at the time of injection. Each animal was evaluated for the extent of label in the duodenal epithelium and if these cells were labeled above the critical value (four grains/nucleus), that animal was not included in the analysis. If the critical value was not exceeded in the duodenal epithelium, isolated cells found elsewhere with a label above this value were considered to represent cells which had been labeled in the thymus and migrated therefrom. Labeled cells identified in various tissues The following is a description of the tissues analyzed and the cells identified. In order not to be repetitive in the use of the term “labeled cell,” this designation is implied wherever an identified cell is mentioned. The criteria for the identification of the several types seen is as follows: 1. Lymphocytes were characterized by their heavily-massed chromatin, distinct nucleoli, and limited basophilic cytoplasm, the latter being relatively more extensive the larger the cell. The distinction between small lymphocytes and the category medium and large was based on nuclear diameter, all cells 5 and below classed as small, and above this as medium and large. 2. The plasma cell series differed in the greater condensation of the nuclear material and increased area and basophilia of the cytoplasm. 3. Macrophages were designated on the basis of engulfed particulate matter, or extensive, usually vacuolated cytoplasm. 4. Reticular cells were distinguished from lymphoid by their delicate chromatin pattern and in their faintly stained cytoplasm. 5. Heterophil myelocytes had polymorphic nuclei and specific granulation. Mesenteric lymph node. The mesenteric lymph node contained the highest number of migrated cells in nearly all animals studied. As seen in table 2, the average number of cells identified in these nodes per 100 oil fields was 13. Although the number varied widely among the several animals, a few labeled cells could always be identified here. At one day a mitotic figure was observed, which was the only such instance in any of the animals of the study. Also seen at this interval were several large, medium, and small lymphocytes, as well as a few macrophages, reticular cells, and one pre-plasma cell (figs. 7, 8, 9). At the two day interval, besides the cell types previously mentioned, early heterophils, with the typical horse-shoe nucleus of the metamyelocyte, and two dead cells engulfed in the cytoplasm of unlabeled macrophages were seen. There was some indication of an increase of cells with time, and in one node at three days as many as 38 cells/ 100 fields were indentified. There was no consistent pattern of distribution, cells Fig. 1 Radioautograph of thymus cortex of a guinea pig which two days previously received 50 pc H3-thymidine locally. Exposure 14 days. Note the extent and intensity of label in small lymphocytes. x 1,800. Fig. 2 Radioautograph of thymus cortex of a guinea pig which 30 minutes previously received 4 pc of H3-thymidine locally. Arrows indicate three large labeled lymphocytes. “a” focused at nuclear level and “b” focused on the silver grains. Exposure 14 days. X 2,100. FATE OF LOCALLY LABELED THYMOCYTES Figures 1 and 2 117 118 RAYMOND G. MURRAY AND PHYLLIS A. WOODS .. 1 (’61) who found after autologous thymus cell transfusion in the guinea pig the opposite trend with time. As in the mesenteric Iymph node, the average grain counts showed no pattern with time, and the cell types observed here were similar to those in the nodes. See figure 10 for a preplasma cell in the spleen. Ceruical l y m p h nodes. The number of cells in the cervical lymph nodes was third compared to other tissues, with an average of 3/100 fields. Both left and right nodes were examined, and although one might expect the node adjacent to the labeled lobe of the thymus to contain more, no significant differences were seen. Here, as in other tissues, more medium than small lymphocytes were identified, and were seen usually in the sinuses and cortex. Bone marrow. Occasionally the bone marrow contained a few labeled cells (fig. 11) and in one case as many as six were recorded in 100 fields. These included a reticular cell and three heterophil myelocytes. See figure 12 for an illustration of a heterophil myelocyte in the bone marrow at one day. Another animal after three days had one heterophil myelocyte and one reticular cell, but most of the remaining animals had no labeled cells. Lung. Labeled cells were rarely observed in the lung. A few medium and small lymphocytes, macrophages, dead cells, and one plasma cell were seen. The lung sometimes had large accumulations of hemosiderin which made the analysis somewhat difficult. This material, however could always be distinguished from the silver grains by its brownish color and tendency to occur in large aggregates. Peyer’s patch. Occasional cells were seen at the one and two day intervals, but none was identified at three days. One animal examined after a one day interval had nine positive cells/100 oil 4 30 T . MIN. 1 DAY 2 DAYS 3 DAYS Fig. 3 Change i n average diameter of labeled nuclei i n locally labeled thymus with time after injection of HJ-thymidine. Each point represents 100 measurements in one animal. 44 lot 30 MIN. 1 DAY 2 DAYS 3 DAYS Fig. 4 Change in percentage of labeled cells i n locally labeled thymus with time after injection of H3-thymidine. Each point represents an animal. being found in the cortex, medullary cords, and sinuses at all intervals. The average grain count over cells varied slightly but showed no clear trend with time. Spleen. The spleen usually was second to the mesenteric lymph node with respect to numbers per 100 fields, but sometimes there were more here than in any other tissue. Usually the great majority was seen in the white pulp, and in one animal at one day they were exclusively in this area, but there was some suggestion of a shift toward the red pulp with time. At three days they were approximately evenly divided between red and white pulp. This differs with the findings of Diderhdm Fig. 5 Radioautograph of basal portions of duodenal crypts in a guinea pig which one day previously received 50 pc of H”-thymidine locally i n the thymus while the vessels were occluded. Note absence of label. Exposure 14 days. X 1,700. Fig. 6 Radioautograph of basal portions of duodenal crypts in a guinea pig which one day previously received 83 pc of H3-thymidine locally in the thymus without vascular occlusion. Note numerous labeled epithelial cells. Exposure 14 days. x 1,700. FATE OF LOCALLY LABELED THYMOCYTES Figures 5 and 6 119 120 RAYMOND G . MURRAY AND PHYLLIS A . WOODS TABLE 2 Quantitation of labeled cells found outside t h e t h y m u s in six representative animals Animal number interval and dose Tissue analyzed 1 Number of labeled cells/ 100 oi! immersion fields Average grain count Range of grain counts 50 Id 41pc mesenteric lymph node spleen bone marrow Peyer’s patch duodenal mucosa 7 8 6 9 14 8 10 9 7 5 5-15 4-30 4-20 4-18 5-8 56 Id 25pc mesenteric lymph node spleen bone marrow 2 2 1 12 13 5 4-20 4-25 5 49 2d 41pc mesenteric lymph node spleen cervical lymph node kidney duodenal mucosa 10 9 8 5 3 5 6 7 5 5 4-6 4-20 4-12 5-6 4-7 52 2d 41uc mesenteric lymph node spleen cervical lymph node duodenal mucosa 15 9 2 4 13 18 20 5 5-10 7-30 20 4-7 51 3d 25pc mesenteric lymph node spleen cervical lymph node 10 8 5 12 12 15 7-25 5-25 5-30 53 3d 50pc mesenteric lymph node spleen cervical lymph node duodenal mucosa 38 2 2 4 9 6 20 4-25 6 20 4-7 1 5 In those tissues not mentioned, no labeled cells were located. fields, including dead cells, lymphocytes, and pre-plasma cells. Lamina propria and submucosa of the duodenum. Several labeled cells were located in the lamina propria and submucosa of the gut, representing 8% of all the identified cells outside the thymus. Cells were seen in the lamina propria of the villi, and occasionally were found between the epithelial cells in the wall of the villus (see fig. 13). More dead cells were identified in the lamina propria than in any other tissue examined. The material suggests that cells were being lost into the gut lumen, although the content of the gut was not examined for a confirmation of this point. Liver. I n spite of extensive examination, no cell was ever identified in the liver of any experimental animal. The absence of cells here does not correlate well with current theories as to the liver’s possible function in recirculation of lymphocytes (Fichtelius, ’63). Moreover, thymus cell transfusions have usually resulted in some cells lodging in the liver, usually at the early intervals (see Fichtelius, ’60, for a review of these studies). It is possible that the finding of transfused thymic cells in the liver is a function of the method of study, reflecting the reticulo-endothelial capacity of this organ. Since this capacity seems not to have been exercised in our experiments, it seems likely that our conditions are more nearly physiological than those obtaining in transfusion experiments. Fig. 7-9 Radioautographs of mesenteric lymph nodes from guinea pigs which had been injected one day before autopsy with 50 pc of H3-thymidine directly into, and confined to, the thymus. In each case “a” is focused on the nucleus and “b” on the grains and exposure was 14 days. Fig. 7 Fig. 8 Fig. 9 Medium lymphocyte. Macrophage. x 2,000. Plasma cell. X 3,000. x 3,000. FATE OF LOCALLY LABELED THYMOCYTES Figures 7-9 121 122 RAYMOND G. MURRAY AND PHYLLIS A. WOODS Kidney. Most of the animals examined revealed no cells in the kidney, with the exception of one at two days and one at three days, which had a few lightly-labeled lymphocytes. Unlabeled thymus. The unlabeled right lobe of the thymus was always examined for labeled cells, and positive cells were occasionally identified within it. There is the possibility of local spreading of thymidine into this tissue, and in some instances, a very light general labeling was noted on the periphery of the tissue, confirming the probability that the label had spread into the lobe. However, in other instances, single, well-labeled cells were identified deep within the tissue, and in these instances were assumed to have migrated from the other lobe. Summary of labeled cells in tissues outside the thymus. Figure 14 summarizes the relative distribution of recovered cells among the several tissues examined. Although the absolute numbers seen were low, a rough calculation is possible to estimate the proportion of the total presumed output of the thymus which is represented by the cells identified. Using Yoffey's figure of 4.7 x 10' cells produced /day in the thymus of the 400 gm guinea pig, divided by two since we injected only one lobe, and assuming a 25% labeling efficiency in the thymus itself, one would expect from 1 to 2 labeled cells/100 oil fields if these cells were evenly distributed throughout the body and remained for 24 hours after migration. A s can be seen from table 2, certain tissues contained much higher concentrations than this figure. Since many areas were not examined, we cannot be certain that the remainder of the cells do not lie in the unsampled regions. Although the selection of tissues €or examination may have introduced some bias, it seems clear that the lymphatic tissues contained the bulk of the cells, which suggests a predilection for these sites. Although we cannot account for all of the presumed daily production of the thymus, neither is it necessary to postulate massive cell death within the thymus to explain our results. Summary of cell types identified outside the thymus. Figure 15 compares the percentages of each cell type among the total of labeled cells found. It can be seen that about 60% of the cells identified were lymphoid, with more cells falling in the medium and large than in the small category. This finding could mean either (1) that small lymphocytes leave the thymus and settle in other tissues where they undergo hypertrophy to medium and large lymphocytes, or ( 2 ) that under normal conditions more medium and large lymphocytes actually leave the thymus. All the transforming cells, that is, those identified as reticular cells, macrophages, plasma cells, or heterophil myelocytes, accounted for 24% of the labeled cells. This lends support to the thesis of Maximow ('09) and others, who have considered the lymphocytes as pluripotential stem cells. It is suggested that because of the more nearly physiological treatment of the cells in this study, as contrasted to transfusion studies, they were allowed to display more fully their potentials for development. DISCUSSION Possible limitations of the technique Reutilization. Use of H3-thymidine as a marker is widely accepted (Cronkite, et al., '60), but there remains the uncertainty as to the degree to which the label may be transferred to other cells, that is, reutilized within the organism. Evidence for reutilization has been reported (Trowell, '57; Speirs et al., '62; Bryant, '63; Rieke, '63). It is necessary, then, to consider whether the labeled cells which we found in various tissues might have acquired their label by this means. Since the Figs. 10-13 Radioautographs of tissues from guinea pigs which had been injected one day before autopsy with 40-50 pc H3-thymidine directly into, and confined to, the thymus. When paired, "a" was focused on the nucleus and "b" on the silver grains. Exposure was 14 days in each case. Fig, 10 Pre-plasma cell in the spleen. X 3,000. Fig. 11 Cell with nucleus resembling a reticular cell, in the bone marrow. X 3,000. Fig. 12 Heterophil myelocyte in the bone marrow. x 3,000. Fig. 13 Part of a villus in the duodenal mucosa. Lower arrow indicates a labeled lymphocyte in the epithelium and the upper arrow a labeled dead cell i n the lamina propria. x 1,300. FATE OF LOCALLY LABELED THYMOCYTES Figures 10-13 123 124 RAYMOND G. MURRAY AND PHYLLIS A. WOODS L A M I N A PROPRI CERVICAL L N MESENTERIC L N LIVER 41 6 0 MEDIUM AND L A R G E LYMPHOCYTE SMALL LYMPHOCYTE U NlDENTlFlED 10 20 30% Fig. 15 The distribution of labeled cells among the several types identified outside the thymus in all animals in which the label was originally confined t o the thymus, and at all intervals from 30 minutes to three days. label is thought to be firmly bound in the nucleus, it presumably cannot be transferred to other cells, except by nuclear breakdown. Bryant postulates such a mechanism in his experiments, and Rieke's results in ascites tumors could be explained on the same basis. In order that such transfer occur in sufficient quantities to label other cells, it would seem that either a massive disintegration supplying large amounts to an area, or intimate association of more limited numbers of dead labeled cells with the recipient nucleus would be required. Even when substantial numbers of labeled dead lymphocytes were injected by Diderholm ('61), no transfer of label to host cells was observed. No evidence of substantial numbers of dying labeled cells was found in our study other than in the injected thymus. Thus, there might have been reutilization within the thymus, and some suggestion that this occurred is found in the relatively slow decrease in grain counts with time. This cell death within the thymus might also result in the release of some marker to the general circulation. That this cannot have occurred to a degree sufficient to account for our finding of labeled cells is shown by the same controls (that is, the absence of label in duodenal epithelial cells) as were utilized to assess the degree of leakage of the original label. The possibility remains that labeled cells may have been engulfed by other cells, thus providing an intimate contact whereby the engulfing cell might be labeled. We have, however, seen only two instances of phagocytosis of labeled cells outside the thymus, and no evidence that the nucleus of the macrophage had thus acquired the label. Mims ('62) introduced H3-thymidine labeled thvmus lymphocytes into cultures and although he noted phagocytosis of labeled cells, none of the phagocytes, even though subsequently undergoing mitosis, acquired the label. It seems highly unlikely, therefore. that any significant portion of the labeled cells we identified were labeled in this manner. Other limitations. In the course of handling and ligating the thymus, and the injection of thymidine, some damage to thymic cells was inevitable. The extent of the effect on the cells, however, must FATE OF LOCALLY LABELED THYMOCYTES certainly be less than occurs in the course of extirpation of the thymus and extraction of the cells for preparation of suspensions for reinjection. This is further indicated by the fact that the spleen and the liver, organs which normally acquire particulate matter injected into the blood stream, acquire the major part of transfused cells (Fichtelius, ’60). In our experiments, on the other hand, labeled cells were never found in the liver, and the concentration was usually higher in mesenteric lymph nodes than in the spleen. The objection can be raised that such high concentrations of H3-thymidine as we used may cause cell damage and death by irradiation. This may, in fact, have contributed to the moderate degree of cell damage noted. However, a large portion of the cells were only moderately labeled, and the death of some cells should not essentially affect the fate of less heavily labeled and presumably undamaged cells. This might explain the fact that cells outside the thymus were, on the whole, less heavily labeled than those within the thymus. It is possible that the extent of migration, and possibly even the degree of subsequent differentiation of thymus cells, might have been reduced by local irradiation damage. This does not affect the significance of those migrated cells we observed, but may mean that the numbers we found represent minimum values. Since the thymus is thought by some to be separated from the general circulation by a “hemato-thymic barrier” (Clark, ’63; Weiss, ’63), the direct introduction of the label may have stimulated the thymus cells in a manner alien to the normal activity of the thymus. Thus the finding of plasma cells of thymic origin might have resulted from such an abnormal stimulation. This does not, however, negate the demonstration that thymus lymphocytes are capable of such transformations if sufficiently stimulated. We must allow also for the possibility that not all the cells labeled in, and migrating from, the thymus are lymphocytes. A few developing myelocytes are always seen in and around the thymus, which might account for our finding of labeled myelocytes outside the thymus. We noted, however, that myelocytes mature within 125 the thymus, accumulating and dying in the Hassd’s bodies, and there is no other evidence, such as circulating immature myelocytes, to indicate that they may be migrating as such. Moreover, the myelocytes identified in the thymus were nearly all of the eosinophil series, while those found outside were invariably heterophil myelocytes. Bearing of our observations on current hypotheses concerning lymphocytes and the thymus The immune reaction is believed to be fundamentally a cellular one (Damashek, ’62), and the lymphocyte is widely conceded to be a key cell in this system. Moreover, the role of the thymus in initiating certain of the immunological capabilities of the organism is under extensive investigation (Good and Gabrielson, ’64). The thymus might exert its influence by supplying iymphoid cells directly, by “seeding” other lymphatic tissues resulting in their further proliferation, or by a humoral factor which stimulates lymphocyte production in lymphatic tissue (see review by Arnason et al., ’62). Migration of cells from the thymus to other sites, although demonstrated indirectly by numerous techniques (see introduction) and directly for a few cells by Nossal (’64), has not been clearly proved for significant numbers of cells. We believe that our results supply this proof, and indicate that lymphatic tissues are the preferred sites for this migration. There is almost no evidence, however, in our material, of the migrating cells proliferating in these locations. Moreover, we do not find them in the germinal centers, where proliferation is most intense. On the contrary, we find transformations, which may be significant in an understanding of the fate of thymic cells and possibly of lymphocytes in general. This lack of proliferation of the migrated cells suggests that at least in the adult guinea pig, seeding of other lymphatic tissues for further proliferation does not play a major role. The existence of a humoral factor which stimulates lymphopoiesis, as proposed by Metcalf (’56) and supported by Osoba and Miller (’63 and ’64), is not precluded by our findings. 126 RAYMOND G . MURRAY AND PHYLLIS A . WOODS That the Iymphocyte may act as a stem cell for other types in the hemopoietic tissues was suggested by Maximow ('09) and supported by numerous authors since (see review by Yoffey and Courtice, '56). Yoffey ('62) has recently emphasized the need in the guinea pig of a source, for example the thymus, to supply large numbers of small stem cells to the bone marrow, which transform there into myeloid cells. We find few labeled cells in the marrow, and although we did note transformations to myelocytes, the extent of this process does not accord with the massive seeding postulated by Yoffey. However, as indicated above, our technique may be reducing appreciably the extent of the normal migration. Transformations of lymphocytes to macrophages and to plasma cells have been frequently reported (macrophages-Bloom, '27; Murray, '47; Fichtelius and Diderholm, '61; and others; and plasma cells -Nossal and Makela, '62; and many others). Although Nossal ('64) saw no transformations in experiments similar to ours, we believe we have demonstrated that thymic lymphocytes do change into other cell types. These are not isolated phenomena, but represent about 24% of the cells identified. These data thus support the direct cellular role of the thymus in the immune response of the organism, but the finding of myelocytes and of many unidentifiable cells (no more than 60% could be reliably classified as typical lymphocytes) indicates that the thymic lymphocytes may have a variety of roles, rather than one limited entirely to immune reactions. Most studies using transfusion of thymocyte suspensions (see Diderholm, '61) report the failure to find transformations to other celI types. Murray and Murray ('64), however, found evidence of transformation toward erythroblasts, myelocytes, and macrophages, after both homologous and isologous injections. The portion of cells transforming was low but this may have been due to the more unphysiological conditions relating to the handling of the cells. Bunting and Huston ('21) believed that a major fraction of lymphocytes is lost from the body into the lumen of the intestine. We have recovered a small number of labeled cells in the duodenal mucosa, but it is not possible from our data to establish how significant these are quantitatively, as we have no measure of the time consumed in such migrations. Since the vast majority of lymphocytes in the mucosa remain unlabeled, it must be that either they remain there for many days, or that the majority of migrating cells come from sources other than the thymus. Further possible applications of the local labeling technique Since it has been shown that cellular migration from the thymus may be studied directly in the young adult animal under nearly physiological conditions, the possibilities of the application of this technique merit attention. In order to compare the normal cell furnishing capacity of the thymus at different stages of maturity, it would be well to study the fetal, neonatal, and older animal with this technique. Since it is known that the guinea pig thymus produces fewer cells with respect to the total lymphoid complex than other animals, such as the rat (Yoffey, 'SO), the technique might be adapted for other species. The possible cellular role of the thymus in immunological mechanisms might be studied by this technique, in order to determine the extent of participation of thymic cells in graft rejections, inflammatory response, and antibody production. Although thymus cells have not proved effective in repopulation of tissues depleted by irradiation, the probability that under some conditions they may act as hemopoietic stem cells suggests the need for additional work in this area. Experiments employing the local labeling technique are in progress in our laboratory to explore further the possibility of a protective roIe of the thymus in irradiation damage. It is suggested that the local labeling technique might be adapted to study the cellular kinetics of other tissues, such as spleen, lymph nodes, appendix, and bursa of Fabricius. With proper attention to technique and controls, this method may well prove to be a useful and important tool for exploring cellular migrations and transformations within the organism. FATE OF LOCALLY LABELED THYMOCYTES ACKNOWLEDGMENT The authors wish to acknowledge the technical assistance of Josephine Grosch. LITERATURE CITED Arnason, B. G., B. D. Bryant and B. H. Waksman 1962 A survey of the thymus and its relation to lymphocytes and immune reactions. Blood, 20: 617-628. Bloom, W. 1927 Transformation of lymphocytes of thoracic duct into polyblasts (macrophages) in tissue culture. Proc. SOC.Exp. Biol. and Med., 24: 567-569. Bierring, F. 1960 Quantitative investigations on the lymphomyeloid system in thymectomized rats. Ciba Foundation Symposium on Haemopoiesis: Cell Production and its Regulation. Little, Brown and Co., Boston, pp. 185198. Bryant, B. J. 1963 Reutilization of lymphocyte DNA by cells of intestinal crypts and regenerating liver. J. Cell Biol., 18: 515-523. Bunting, C. H.,and J. Huston 1921 Fate of the lymphocyte. J. Exp. Med., 33: 593-600. Clark, S. L.,Jr. 1963 The thymus in mice of strain 129J. Studies with electron microscopy. Am. J. Anat., 112: 1-33. Cronkite, E. P., V. P. Bond, T. M. Fleidner and S. A. Killman 1960 The use of tritiated thymidine in the study of haemopoietic cell proliferation. Ciba Foundation Symposium on Haemopoiesis: Cell Production and its Regulation. Little, Brown and Co., Boston, pp. 70-92. Damashek, W. 1962 The thymus and lymphoid prdiferation. Blood, 20: 629-632. Diderholm, H. 1961 Studies on the migration and transformation of lymphocytes in immunized and non-immunized animals. Acta Path. Microbiol. Scand., Suppl. 146. Ernstrom, U. 1963 Thyroxin-induced changes in cell composition of lymph node tissues, spleen, and thymus. A quantitative examination of young guinea pigs. Acta Path. Microbiol. Scand., 59: 145-155. Fichtelius, K. E. 1960 On the destination of thymus lymphocytes. Ciba Foundation Symposium on Haemopoiesis: Cell Production and its -Regulation. Little. Brown and Co., Boston, pp. 20'i-220. 1963 Recirculation of lymphocytes via liver and its lymph. Acta Haemat., 29: 233-241. Fichtelius, K. E,., and H. Diderholm 1961 Autoradiographic analysis of the accumulation of lymphocytes in wounds. Acta Path. Microbiol. Scand., 52: 11-18. Good, R. A., and A. E. Gabrielson, eds. 1964 The Thymus in Immunobiology. Harper and Row, New York. Gyllensten, L. 1953 Influence of thymus and thyroid on the postnatal growth of the lymphatic tissue i n guinea pigs. Acta Anat. Suppl., 18: 1-163. Jankovic. B. D.. B. H. Waksman and B. G. Arnason 1962 Role of the thymus i n immune reactions in rats. I. The immunologic response to bovine serum albumin (antibody formation, Arthus reactivity, and delayed ~ 127 hypersensitivity) in rats thymectomized or splenectomized at various times after birth. J. Exp. Med., 116: 159-176. Keohane, K. W., and W. K. Metcalf 1958 Some experiments in fluorescent microscopy designed to elucidate the fate of the lymphocyte. Quart. J. of Exp. Phys., 43: 4 0 8 4 1 8 . Kindred, J. E. 1942 A quantitative study of the hemopoietic organs of young adult albino rats. Am. J. Anat., 71: 207-243. Leblond, C. P. 1959 Classical techniques for the study of the kinetics of cellular proliferation. Kinetics of Cellular Proliferation. Ed. by F. Stohlman, Jr., Grune and Stratton Co., New York, pp. 31-47. Leblond, C. P., and G. Sainte-Marie 1960 Models for lymphocyte and plasmocyte formation. Ciba Foundation Symposium on Haemopoiesis: Cell Production and its Regulation. Little, Brown and Co., Boston, pp. 152-172. Maximow, A. 1909 Untersuchungen ueber Blut and Bindegewebe. I. Die Fruehesten Entwicklungsstadien der Blut - und Bindegewebszellen beim Saeugetierembryo, bis zum Anfang der Blutbildung in der Leber. Arch. Mikr. Anat., 73: 444-561. Metcalf, D. 1956 The thymic origin of the plasma lymphocytosis stimulating factor. Brit. J. Cancer., 10: 442457. 1960 The effect of thymectomy on the lymphoid tissues of the mouse. Brit. J. Haemat., 6: 324. Miller, J. F. A. P. 1963 Role of the thymus in immunity. Brit. Med. J., 2: 459-464. Miller, J. F. A. P.,A. H. E. Marshall and R. G. White 1962 The immunological significance of the thymus. Adv. in Immunol., 2: 111-162. Mims, C. A. 1962 Experiments on the origin and fate of lymphocytes. Brit. J. Exp. Path., 43: 639-649. Murray, R. G. 1947 Pure cultures of rabbit thymus epithelium. Am. J. Anat., 81: 469-511. 1962 Fate of thymidine-H3 labeled lymphocytes after intravenous and subcutaneous injections in rats and guinea pigs. Anat. Rec., 142: 261 (abstract). Murray, R. G . , and A. Murray 1961 Studies on the fate of lymphocytes. I. Labeling small thymic lymphocytes with tritiated thymidine. Blood, 18: 737-749. 1964 Studies on the fate of lymphocytes. 11. Intravenous injection of labeled thymic lymphocytes into homologous rats and isologous mice. Anat. Rec., 150: 95-112. Nossal, G. J. V. 1964 Studies on the rate of seeding of lymphocytes from the intact guineapig thymus. From 6th Int. Transplantation Conf., ( i n press). Nossal, G. J. V., and 0. Makela 1962 Autoradiographic studies on the immune response. I. The kinetics of the plasma cell proliferation. J. Exp. Med., 115: 209-230. Osoba, D., and J. F. A. P. Miller 1963 Evidence for a humoral thymus factor responsible for the maturation of immunological faculty. Nature, 199: 653-654. 1964 The lymphoid tissues and immune responses of neonatally thymectomized - 128 RAYMOND G . MURRAY AND PHYLLIS A. WOODS mice bearing thymus tissue in millipore diffusion chambers. J. Exp. Med., 119: 177-194. Papermaster, B. W., R. M. Condie, J. Finstad and R. A. Good 1964 Significance of the thymus in the evolution of the lymphoid tissue and acquired immunity. The Thymus in Immunobiology. Harper and Row, New York, pp. 551591. Quastler, H., and F. G . Sherman 1959 Cell population kinetics in the intestinal epithelium of the mouse. Exp. Cell Res., 17: 420-438. Rankin, J. J. 1960 Circulation of lymphocytes and thymocytes in the normal and leukemic mouse. Acta Anat., 40: 221-230. Rieke, W. 0. 1962 The in wino reutilization of lymphocytic and sarcoma DNA by cells growing in the peritoneal cavity. J. Cell Biol., 13: 205216. Sainte-Marie, G., and C. P. Leblond 1958 Origin and fate of cells in the medulla of rat thymus. Proc. Exp. Biol. and Med., 98: 909-915. 1964 Cytological features and cellular migration in the cortex and medulla of thymus in the young addult rat. Blood, 23: 275299. Speirs, R. S., E. E. Jansen, S. Speirs, S. Osada and and L. Dienes 1962 Use of tritiated thymidine to study the origin and fate of inflamma- tory cells. Tritium in the Physical and Biological Sciences. Vol. 11. Int. At. Energy Comm., Paul Gerin, Vienna, Austria, pp. 301-325. Trowell, 0. A. 1957 Reutilization of lymphocytes i n lymphopoiesis. J. Biophys., and Biochem. Cyt., 3: 317-318. Walker, B. E., and C. P. Leblond 1958 Sites of nucleic acid synthesis in the mouse visualized by radio-autography after administration of C14 labelled adenine and thymidine. Exp. Cell Res., 14: 510-531. Weiss, L. 1963 Electron microscopic observations on the vascular barrier in the cortex of the thymus of the mouse. Anat. Rec., 145: 413-438. Woods, P. A. 1964 Local labeling of thymocytes in the guinea pig with tritiated thymidine. Anat. Rec., 148: 352 (abstract). Yoffey, J. M. 1960 Quantitative Cellular Haematology. Charles C Thomas, Springfield, Illinois. 1962 The present state of the lymphocyte problem. Lancet, 1: 206-211. Yoffey, J. M., and F. C. Courtice 1956 Lymphatics, Lymph, and Lymphoid Tissue. Harvard Univ. Press, Cambridge, Mass. Yoffey, J. M., W. 0. Reinhardt and N. B. Everett 1961 The uptake of tritium-labelled thymidine by lymphoid tissue. J. Anat., 95: 293-299.