Relationship of Adenovirus to Lymphocytes in

Transcription

1 INFCTION AND IMMUNITY, Apr. 973, p American Society for Microbiology Vol. 7, No. 4 Printed in U.S.A. Relationship of Adenovirus to Lymphocytes in Naturally Infected Human Tonsils and Adenoids J. VAN DER VEEN AND M. LAMBRIEX Laboratory of Bacteriology, Immunology anid Virology, St. Elisabeth Hospital, Tilburg, Received for publication 7 January 973 Purified lymphocytes from human tonsil and adenoid specimens were cultured with and without phytohemagglutinin. Adenovirus was isolated from lymphocytes of 8 of 90 specimens tested. With one exception, it was necessary to culture the lymphocytes before infectious virus could be detected. Phytohemagglutinin stimulation enhanced the recovery of virus. The results suggest that lymphocytes in tonsils and adenoids may be naturally infected with adenovirus and that, in positive cultures, at least of every 07 cells harbors virus or viral precursor at initiation of the cultures. Adenovirus was demonstrated directly in fresh suspensions of unpurified cells from tonsils and adenoids in seven cases. In five of these cases, at least of every 06 cells contained infectious virus. Adenovirus was isolated from 6 (62%) of 98 tonsil and adenoid specimens by the conventional method of tissue fragment culture after various periods of cultivation. The viruses isolated were of serotypes, 2, 5, and 6. The frequent association of adenovirus with human tonsils and adenoids has long been known. The viruses isolated almost invariably are of serotypes, 2, 5, and 6 (references 4, 5, 0, ). Virus is present in these tissues without evidence of clinical illness and can be revealed by growing tissue fragments in vitro. Apparently, adenovirus may persist in tonsils and adenoids as a latent infection. There have been few investigations on the relationship of adenovirus to tonsil and adenoids at the cellular level. In cultures of adenoids and tonsils, cytopathic effects (CPE) characteristic of adenovirus are observed in epithelial and fibroblastic cells (4, 0). Strohl and Schlesinger (2, 3) studied quantitative aspects of the natural infection. They suggested that, among the various cell types making up a primary suspension, most likely only fibroblasts were susceptible to viral infection. The present study was undertaken in an effort to determine whether lymphocytes from tonsils and adenoids harbor adenovirus and contribute to the maintenance of viral infection in these tissues. Fresh suspensions of lymphocytes as well as phytohemagglutinin (PHA)-treated anid untreated lymphocyte cultures were studied. For comparison, virus was sought also in unpurified cell cultures and in fragment cultures of tonisils and adenoids. T'he Netherlands MATERIALS AND METHODS Preparation of tissue fragments and unpurified cell suspensions. Whole tonsils and adenoids removed from children between and 8 years of age were transported in cold Hanks solution and were processed within h. The majority of tonsils and adenoids used were markedly enlarged. The tissues were trimmed of blood clots and mucus. Unwashed tissues were minced into very small fragments by transferring them to a stainless steel tube fitted with a steel plunger and a sieve with a pore diameter of approximately 0.75 mm (Harrison, London). Tissues were forced through the sieve by screwing the plunger. Fragments were collected in 0 ml of culture medium consisting of medium 99 and 20% calf serum and were shaken for 5 s in a Vortex mixer. The cells released from the fragments were drawn off with the supernatant fluid and kept at 36 C. Fragments were resuspended in 0 ml of culture medium anid were shaken again in the same manner. The procedure was repeated three to five times. The supernatant cell suspensions were pooled. They are referred to as "unpurified cells." The yield of viable cells recovered in this manner ranged from 200 X 06f to,300 X 06 cells per adenoid or pair of tonsils (average 600 X 06 cells). Fragments were washed with Hanks solution to remove adhering cells. The supernatant fluids containing these cells were discarded. The procedure was repeated until the supernatant fluid became clear. Preparation of lymphocyte cultures. The unpurified cell suspensions were used as source of lymphocytes. The procedure employed was essentially the same as that described by Lamvik (6). 604

2 VOL. 7, 973 ADENOVIRUS AND LYMPHOCYTES 605 An 8-inch (20.32-cm), water-j acketed glass column, 0.75 inch (9 mm) in diameter, was loosely packed With nylon fibers, connected to a 36 C water bath, and rinsed with Hanks solution. Subsequently, cell suspension prewarmed at 36 C was passed onto the column. After an adsorption period of 45 min the cells were eluted; the flow was adjusted to 0 drops per min. The first few milliliters of effluent was again delivered to the column. After the cell suspension had been applied to the column, 75 ml of Hanks solution with 20% calf serum was added to the column to elute additional lymphocytes. The cells in the effluent were washed three times with Hanks solution and resuspended in culture medium at a concentration of.5 X 06 lymphocytes per ml. The yield of viable lymphocytes ranged from 00 X 06 to 450 X 06 cells per adenoid or pair of tonsils (average 20 X 06 cells). Lyophilized PHA-M (Difco Laboratories, Detroit, Mich.) was dissolved in 5 ml of phosphate-buffered saline (PBS) and added to the lymphocyte suspension at a final concentration of 0.5% unless otherwise indicated. Control cultures of lymphocytes received no PHA. PHA-treated and untreated lymphocytes were distributed to 6- by 25-mm screw-cap glass culture tubes in samples of 2 ml per tube. Tubes were incubated upright at 36 C. Fifty percent of the medium was replaced after and 4 days of incubation. Cultures were harvested after 8 days. Portions of supernatant fluids were frozen at -70 C. All cultures were pooled for each specimen individually, washed three times with Hanks solution, and resuspended in Hanks solution. This pooled suspension was subjected to six cycles of freezing and thawing. The harvest containing supernatant fluid and cell debris was stored at -20 C until it was used for virus isolation. Freshly prepared lymphocyte and unpurified cell suspensions were also assayed for virus. The procedure employed in harvesting these cells was the same as that used for cultured lymphocytes. The total number of viable cells in the pooled suspensions ranged from 6 X 06 to 72 X 06 for lymphocytes and from 0.75 X 06 to 36 X 06 for unpurified cells Cy tomorphology and cell viability. Differential counts of cell suspensions collected before and after processing on the column and of cell cultures at various times of incubation were made by use of Giemsa-stained preparations. A total of,000 cells from each of three cultures were examined. Identification of "blast-like" cells was based on the morphological criteria of Chessin et al. (2), and these cells will be referred to as blast cells to distinguish them from small lymphocytes. Cell viability was estimated with trypan blue dye, and cell counts were obtained with a hemocytometer. Samples of each of three cultures of each series were counted in duplicate. Measurement of 3H-thymidine incorporation into DNA. Five hours prior to termination of culture,.0 ;&Ci of 3H-thymidine (Radiochemical Centre, Amersham, England; specific activity 5 Ci/mmol) was added to each of three culture tubes. At the time of harvesting, the tubes were placed in an ice bath for 0 min. Cells were sedimented by centrifugation at,000 rpm for 5 min and washed once in ice-cold PBS. Cold trichloroacetic acid (5%) was added, and the acid-insoluble precipitate was sedimented at 3,000 rpm for 0 min. The sediment was dissolved in ml of 0. mm sodium hydroxide and again precipitated with 4.5 ml of cold trichloroacetic acid (6.7%). The resulting precipitate was again sedimented by centrifugation, and.0 ml of ether-acetone mixture was added to the precipitate. The ether-acetone mixture was evaporated in a 40 C water bath for 5 min. The dry precipitate was dissolved in 0. ml of hyamine 0-X (Packard Instrument Ltd., London) and suspended in 2.5 ml of scintillation fluid. The radioactivity was counted in a Packard Tri-Carb automatic scintillation counter (model 3380). Results are expressed as the mean counts per minute of triplicate cultures. Fragment cultures. Screw-cap glass tubes which were to receive tissue fragments were wetted with inactivated calf serum to promote sticking of the tissue to the surface of the glass. Fragments were transferred with a bent-tip Pasteur pipette to one side of these tubes, 300 to 400 fragments per tube; six to eight tubes were used per pair of tonsils or adenoid. One milliliter of growth medium, consisting of 20% inactivated calf serum and 0.03% glutamine in Eagle minimal essential medium, was added per tube. The tubes were placed in an incubator at 36 C in an inverted position at an angle of 50 to the horizontal so that the fragments were on the upper side and the medium on the lower side. After 8 h the tubes were turned 800 and incubated in a stationary state. Medium was replaced 24 h after initiation of incubation. Thereafter, all tubes were incubated alternately in a stationary state for 0 h and in a roller drum for 4 h. Medium was changed every second or third day. Cultures showing CPE characteristic of adenovirus were subjected to six cycles of freezing and thawing. The harvest containing fluid and cell debris was used for virus isolation. Cultures which gave no CPE were maintained for 0 weeks and then treated in the same manner. Virus isolation. Primary human thyroid cultures (4) were employed for virus isolation. A 0.2-ml amount of material to be tested for virus was inoculated into each of two tube cultures. At least two passages corresponding with a total incubation period of 30 days were made before being discarded as negative. All viral strains isolated were typed by means of neutralization tests with rabbit antisera against adenovirus prototype strains (5). RESULTS Composition of cell preparations. Ini freshly prepared unpurified cell suspensions, 95 to 98% of the cells were lymphocytes, including those of small and medium size as well as large blast cells. The remaining 2 to 5% comprised

3 6006 VAN DER VEEN AND LAMBRIEX INFECT. IMMUNITY phagocytic cells (monocytes and polymorphonuclear leukocytes) and occasional epithelial cells. Strohl and Schlesinger (3), utilizinig cover-slip cultures with known numbers of unpurified cells after cultivation for various periods of time, estimated that approximately 0.004% of these cells were fibroblasts. In fresh preparations and cultures of column-purified cells, lymphocytes were noted exclusively. Cell differentials performed on as many as 0,000 cells of each of three purified lymphocyte preparations likewise failed to reveal the presence of contaminating cells. To check further the purity of the lymphocyte preparation, tube cultures of column-purified lymphocytes were incubated at 36 C in stationary racks in a horizontal position for 8 days. None of duplicate lymphocyte cultures from 70 adenoids showed cells which adhered to the glass. There was no decrease in ph. In contrast, stationary tube cultures of unpurified cell preparations demonstrated areas of glass-adherent cells (presumably macrophages) with lymphocytes attached to these cells on the second day of culture. Additional areas of fibroblastic cells appeared after approximately 5 days of culture. Metabolic activity could be recognized grossly by the fairly sharp drop in ph. Response of lymphocytes to PHA. Freshly prepared cultures of tonsillar and adenoid lymphocytes were stimulated with different concentrations of PHA and harvested at daily intervals after a 5-h pulse with 'H-thymidine. There was considerable variation in the degree of thymidine incorporation between lymphocytes from different tissues. The peak of activity occurred between the 3rd and 5th days of culture. Maximal response was observed at PHA concentrations of 0.5 and.0% (Fig. ). Since the total number of viable lymphocytes on days 4 and 8 of culture were always highest at a PHA concentration of 0.5%, this concentrationl was chosen for further experiments. Unstimulated lymphocyte cultures showed a reduced but still significant degree of thymidine incorporation (2,000-5,000 counts/ min), which remained constant during at least 5 days. August et al. () found a high rate of thymidine incorporation (approximately 20,000 counts/min) by unstimulated tonsillar lymphocyte cultures at 24 h after initiation of the cultures, but this declined gradually during the following 6 days. Determination of the number of blast cells in fresh and 4-day-old cultures (Table ) shows that blast transformation occurs in unstimulated cultures. Superimposed upon this transformation is the response to PHA. However, in cultures already highly stimulated naturally, as indicated by high counts of blast cells in fresh preparacpm x 07 60] o % PHA no.ofcells /ml x05, FIG.. Dose-response curves for PHA showing incorporation of 3H-thymidine (0) and number of viable cells (0) on day 4 for lymphocyte cultures from three adenoids. TABLE. Blast cell transformation in PHA -treated and untreated lymphocyte cultures from adenoids Increase (+) or decrease (-) in blast cells/ No. of blast cells on day 4a ml on day specim X 06 mens Untreated PHA-treated (%) % 0.78 or< (43-09) +00 (47-95) or> 7-45 (30-63) -23 (0-47) a Values are the mean increases or decreases and those in parenthesis are the range for different specimens. tions, no further response was noted by morphological criteria. Oni the contrary, the number of blast cells was decreased by the 4th day in these cultures. Isolation of adenovirus from lymphocytes and unpurified cells. Adenovirus isolations were attempted from fresh lymphocyte preparations and from PHA-treated and untreated 8- day-old lymphocyte cultures of 90 tissue specimens (27 tonsils aind 63 adenoids). In addition, of 58 (3 tonsils and 55 adenoids) of these specimens freshly prepared, unpurified cell suspensions were tested for the presence of virus. Cells were washed aind disrupted by freezing and thawing. This material was inoculated into primary human thyroid cultures as described under Materials and Methods. In this way adenovirus was demonstrated in cells from specimens (Table 2).

4 VO0L. 7, 973 ADENOVIRUS AND LYMPHOCYTES 607 TABLE 2. Adenovirus isolation from fresh and cultured lymphocytes with and without PHA and from fresh utnpurified cells of 27 tonsils and 63 adenoids Composition of cultures Purified lynmphocytes Fresh Cultured, -PHA Cultured, +PHA Unpurified cells Fresh Adenovirus type recovered Adenovirus-positive specimens Tonsils A + B C D Adenoids E F G!H + I JJJK 2 5 Lymphocyte preparations yielded adenovirus in eight instances. The total number of lymphocytes tested for the presence of virus per viruspositive specimen ranged from 6 X 06 to 60 X 06. With one exception it was necessary to culture the lymphocytes before virus could be detected. Stimulation by PHA further enhanced the recovery of virus. In one case infectious virus was demonstrated in uincultured lymphocytes. The adenoid from which these lymphocytes were derived was probably heavily iinfected with virus, since fragment cultures of this tissue showed CPE as early as 2 days after initiation of the cultures. It is noteworthy that in seven inistances adeinovirus was isolated directly from freshly prepared, unpuried cell susp)ensions. The total numbers of cells tested for the presence of virus per viruspositive suspensioin were (see Table 2): F, 0.75 X 06; E, H, J and K,.5 X 06; I, 3 X 06; and C, 3 X 06. The positive suspensions were assayed for virus by the tube dilution method. One suspeinsion was positive at a dilution of /0. The remaining six suspensionis were positive only after inoculation of undiluted material. These finldings would imply that cells containinig an amount of infectious virus large enough to be directly recoverable are present in naturally infected tissues. Column purification would seem to result in loss of these cells. Isolation of adenovirus from tissue fragments. Tonsil and adenoid tissues grew readily. Adenoids showed initially ani outgrowth of epithelioid cells. Older cultures consisted of fibroblastic cells and areas of epithelioid cells. In the case of tonsils, there was predominantly fibroblastic outgrowth. In all, 8 of the 06 tissues cultured had to be discarded because of contamination with bacteria or yeasts. Table 3 shows that adenovirus was isolated from 20 (57%) of 35 pairs of tonsils and from 4 (65%) of 63 adenoids. The adenoviruses recovered were of types, 2, 5, or 6, in this order of frequency. No double infections were found. The time of first appearance of CPE was usually between 5 and 5 days after initiation of the cultures. In all instances in which adenovirus was recovered from lymphocyte or unpurified cell cultures, the corresponding fragment cultures were also positive and the same adenovirus types were found in cell and fragment cultures. DISCUSSION The present studies provide suggestive evidenice that lymphocytes in naturally infected tonisils or adenoids represent one of the cell types which may harbor adenovirus or viral precursor. Purified lymphocyte cultures from 8 of 90 tissues studied were positive. It must be admitted that the identification of the cell type in these cultures which yields virus is only tentative, since the presence of contaminating cells cannot be excluded with certainty. However, several findings are consistent with the suggestion that it is the lymphocyte. Smears of purified cultures revealed exclusively lymphocytes and never other cell types. No glass-adherent cells were found in tube cultures incubated horizontally for 8 days, confirming the purity of the cultures. With the exception of one tissue, adenovirus was recovered only from cultured lymphocytes, not from fresh ones. Thus, it may be assumed that either the amount of virus supposed to be initially present in lymphocyte preparations is very small or fresh lymphocytes contain only viral precursor. In either case viral replication is required before virus can be detected. These data appear to rule out the possibility that adenovirus recovered from lymphocyte cultures merely represented surviving virus adsorbed to cells. PHA stimulation of lymphocytes enhanced the T.tBLE: 3. Adenovirus isolation from fragmenzt cultures of tonsils and adenoids No. of No. of adenovirus strains isolated Tissue 8peciseiesmens Cul- Total Type Type Type Type secimens tured Tonsils Adenoids

5 608 VAN DER VEEN AND LAMBRIEX INFECT. IMMUNGITY recovery of virus. This finding might be related to the enhancing effect of PHA on replication of various viruses in lymphocyte cultures (3, 6). It is to be noted, however, that adenovirus was also isolated from untreated lymphocyte cultures. The present investigation and studies of others (, 8) indicate that there is a considerable degree of background synthetic activity in untreated cultures of tonsillar and adenoid lymphocytes. Since tonsils and adenoids are exposed continuously to different kinds of antigens, it may be assumed that this reflects the effect of stimulation initiated in vivo. The recovery of adenovirus from untreated lymphocyte cultures may then be explained by assuming that lymphocytes stimulated by naturally occurring specific antigens are capable of supporting virus growth in a similar manner, though less effective, as do PHAstimulated lymphocytes. A more direct approach to this question is the study of exogenous adenovirus infection of PHA-stimulated and unstimulated lymphocyte cultures. This will be the subject of a separate study. Lymphocyte cultures yielded adenovirus less often than did fragment cultures. This might be related to the number of cells involved. A total of 6 X 06 to 60 X 0 lymphocytes per tissue were tested for the presence of virus. It is possible that tests with greater numbers of lymphocytes would have revealed virus in more specimens. Strohl and Schlesinger (3) utilizing trypsin-dispersed cell suspensions from tonsils or adenoids, found suggestive evidence that only in 07 to 09 cells harbors virus or viral precursor. It is also possible that qualitative differences in cell composition between lymphocyte and fragment cultures are responsible for the low frequency of positive lymphocyte cultures, assuming that the infection is located more frequently in cells other than lymphocytes. The possible role of each of the various cell types present in tonsils and adenoids in maintaining adenovirus infection in vivo in these tissues has not yet been clearly defined. The susceptibility of epithelial cells and fibroblasts in tonsil and adenoid tissues grown in vitro has long been recognized (4, 0). Strohl and Schlesinger (2, 3) reported extensive studies on the possible role of fibroblasts in the persistence of adenovirus infection. They suggest that fibroblasts might represent the major, if not the only, susceptible cell type in primary cell suspensions prepared from tonsils or adenoids, implying that lymphocytic cells were refractory to adenovirus. The discrepancy with our results can be easily explained since the methods applied by these authors were inadequate for cultivating lymphocytes. The finding in the present study that freshly prepared unpurified cell suspensions were a better source of infectious virus than were fresh purified lymphocyte preparations would seem to suggest that the lymphocyte is not the principle cell type which determines the persistence of inlfection in vivo. It is conceivable that the lymphocyte only plays a role in initiating infection of tonsils and adenoids. Lymphocytes infected with virus at the primary site of infection would enter into the tonsillar and adenoid tissues. Virus would then be passed from these cells to other cell types. The continuous presence of a small number of infected lymphocytes in tonsils and adenoids might be explained by assuming that some cells may carry virus or viral precursor for a considerable time. Another possibility is that lymphocytes constituted the target of adenovirus produced by tonsillar and adenoid cells of a different type of types. Studies of Nasz et al. (7), employing the immunofluorescent antibody technique, favor the hypothesis that adenovirus or adenovirus antigen may be present in lymphocytes, at least temporarily. The high recovery rate of adenovirus from fragment cultures of tonsils and adenoids found in the present investigation accords well with previous studies (4, 5, 9-, 3). Characteristic CPE was detected soon after initiation of incubation of the cultures as compared with the results of previous studies. The reason for this early CPE is presumably that a large amount of tissue of each specimen was tested (4) and that the tissue was minced into very small fragments with a sieve, allowing rapid release and outgrowth of cells from the fragments. A noteworthy finding is the direct recovery of adenovirus from freshly prepared, unpurified cell suspensions in seven instances. This finding is in contrast to the commonly held view that the presence of adenovirus in naturally infected tonsils and adenoids can be revealed only by growing explanted tissue in vitro (4). Previous reports (0, 3) have described only two instances in which adenovirus was recovered directly from fresh tonsils or adenoids, and in one of these the recovery was presumably associated with acute febrile illness. The discrepancy between our findings and those of previous studies is difficult to explain. Rowe et al. (0) have suggested that the presence of neutralizing antibody might be responsible for difficulty in recovering virus from fresh tissues. It is possible that our procedure of washing eells was effective in removing antibody. The study of purified preparations of each of the different cell types present in tonsillar and adenoid tissues may lead to thc identification of the host cells producing infectious virus in vivo.

6 VOL. 7, 973 ADENOVIRUS AND LYMPHOCYTES 609 ACKNOWLEDGMENTS We thank J. J. W. Roovers of the Maria Hospital, Tilburg, for providing us with specimens of tonsils and adenoids, and F. C. J. C. Soontielns and C. C. Raymakers-Volaart for their help in assaying lymphocyte stimulation by 3H-thymidine incorporation. LITERATURE CITED. August, C. S., E. Merler, D. 0. Lucas. and C. A. Janeway The response in vitro of human lymphocytes to phytohemagglutinin and to antigens after fractionation on discontinuous density gradients of albumin. Cell. Immunol. : Chessin, L. N., J. Borjeson, P. D. Welsh, S. D. Douglas, and H. L. Cooper Studies on human peripheral blood lymphocytes in vitro. II. Morphological and biochemical studies on the transformation of lymphocytes by pokeweed mitogen. J. Exp. Med. UL Eustatia, J. M., and J. van der Veen. 97. Viral replication in cultures of phytohemagglutinintreated mouse lymphocytes. Proc. Soc. Exp. Biol. Med. 37: Evans, A. S Latent adenovirus infections of the human respiratory tract. Amer. J. Hyg. 67: Israel, M. S The viral flora of enlarged tonsils and adenoids. J. Pathol. Bacteriol. 84: Lamvik. J Separation of lymphocytes from human blood. Acta Haematol. 35: Nasz, I., G. Kulcs8r, P. Din, and K. Sallay. 97. A possible pathogenic role for virus-carrier lymphocytes. J. Infect. Dis. 24: Oetggen, H. F., R. Silber, P. A. Miescher, and K. Hirschhorn Stimulation of human tonsillar lymphocytes in vitro. Clin. Exp. Immunol. : Rowe, W. P., R. J. Huebner, L. K. Gilmore, R. H. Parrott, and T. G. Ward Isolation of a cytopathogenic agent from human adenoids undergoing spontaneous degeneration in tissue culture. Proc. Soc. Exp. Biol. Med. 84: Rowe, W. P., R. J. Huebner, J. W. Hartley, T. G. Ward, and R. H. Parrott Studies of the adenoidal-pharyngeal-conjunctival (APC) group of viruses. Amer. J. Hyg. 6: Schlesinger, R. W. 96. Vagaries of adenoviruscell complexes, p. 69. In M. Pollard (ed.), Perspectives in virology, vol. 2. Burgess, Minneapolis. 2. Strohl, W. A., and R. W. Schlesinger Quantitative studies of natural and experimental adenovirus infections of human cells. I. Characteristics of viral multiplication in fibroblasts derived by long-term culture from tonsils. Virology 26: Strohl, W. A., and R. W. Schlesinger Quantitative studies of natural and experimental adenovirus infections of human cells. II. Primary cultures and the possible role of asynchronous viral multiplication in the maintenance of infection. Virology 26: van der Veen, J., and J. H. Dijkman Association of type 2 adenovirus with acute respiratory illness in military recruits. Amer. J. Hyg. 76: van der Veen, J., and G. Kok Isolation and typing of adenoviruses recovered from military recruits with acute respiratory disease in the Netherlands. Amer. J. Hyg. 65: Wheelock, E. F., S. T. Toy, and R. L. Stjernholm. 97. Interaction of viruses with human lymphocytes, p In D. B. Amos (ed.), Progress in immunology, Academic Press Inc., New York.