m. J. Trop. Med. Hyg., 59(1), 1998, pp. 166 170 opyright 1998 by The merican Society of Tropical Medicine and Hygiene TRYPNOSOM RUZI IN THE RETUM OF THE UG TRITOM INFESTNS: EFFETS OF LOO INGESTION Y THE STRVE VETOR STRI H. KOLLIEN, and GÜNTER. SHU epartment of Special Zoology, Ruhr University, ochum, Germany bstract. To follow the developmental effects of feeding of the insect host after long starvation periods, the population density and composition of an established infection of Trypanosoma cruzi in the rectum of Triatoma infestans were determined 60 days after the last feeding (daf) and then at different intervals after feeding. The original population decreased and then increased up to the 10th daf. In starved bugs, about 30% were spheromastigotes (including intermediate forms), 20% epimastigotes, and 50% trypomastigotes, but one daf, these forms represented 2%, 70%, and 10%, respectively. In addition, one daf there were about 10% giant cells, i.e., a multiple division stage. In the following two days, this form represented on average 30 50% of the total population, but it then disappeared nearly completely. Thus, giant cells evidently develop by rapid growth of epimastigotes, if conditions become optimal after long starvation periods of the vector. Trypanosoma cruzi, the etiologic agent of hagas disease, is transmitted by triatomine bugs. In the bug s intestinal tract, many different developmental stages of T. cruzi appear, as had already been drawn in the original description of the species by hagas. 1 eside the three main stages (epimastigotes, trypomastigotes, and spheromastigotes), a variety of intermediate stages develop that can be summarized as flagellates with either a drop-like shape (intermediates between spheromastigotes and epimastigotes or trypomastigotes) or a slender shape (intermediates from epimastigotes to trypomastigotes). 2 The biological importance is known only for some stages. The metacyclic trypomastigotes develop in the rectum and are infectious for mammals. Epimastigotes are, in contrast to trypomastigotes, able to multiply and colonize the whole intestinal tract of the vector. oth stages dominate in an established T. cruzi population of regularly fed bugs, in which spheromastigotes rarely occur. Their percentage increases in the rectum of starved bugs as an adaptation of the flagellate to stress conditions. 3 n investigation of the T. cruzi population after the change of the nutritional situation by feeding starved bugs offers the possibility to clarify the further development of the spheromastigotes and drop-like stages. by the molt to the fifth instar within 18 days after feeding (daf). Ten bugs were dissected after a starvation period of 60 days; other nymphs were fed on hens and 10 L5 for each time point were dissected (1, 2, 3, 5, and 10 daf). The population density was determined for the rectal wall and the rectal lumen using Neubauer hemocytometers. Smears were made of each sample 8 and after Giemsa staining more than 150 flagellates were classified into 19 different forms, which were then summarized into the three main groups: spheromastigotes, epimastigotes, trypomastigotes, as well as two intermediate forms (between spheromastigotes and epimastigotes or trypomastigotes and between epimastigotes and trypomastigotes) 2 and multiple division forms (giant cells). This classification was impossible at very low population densities. two-factor analysis of variance (NOV) was calculated using the software of SS Institute, Inc. (ary, N) and a significant NOV was followed either by a uncan s multiple range test 9 for one significant factor or by the Student s t-test if the combination of the two factors was significantly different. RESULTS MTERILS N METHOS The strain of T. cruzi (hile 5) was isolated from Triatoma infestans that originated from achiyuyu. 4 It was maintained cyclically between mice and bugs or stored frozen at 78. 5 Triatoma infestans strain hile originated from the same locality in hile. Three separate stocks were initiated in 1979, 6 and all were mixed in 1991. The bugs were fed on hens and reared at 26 1 and 60 70% relative humidity on a 16 hr : 8 hr light : dark photoperiod. 7 First instar nymphs (L1) of Triatoma infestans were fed on T. cruzi-infected mice (about 3 10 6 blood trypanosomes/ml). Groups of about 150 fully engorged L1 were maintained in two-liter beakers. 2 In vitro cultured symbionts isolated from a domestic population of Triatoma infestans (Reintjes N, Schaub G, unpublished data) were added after feeding. ugs were fed on hens every three weeks, finishing 13 weeks postinfection (pi) with L4. Only totally engorged insects were selected; the feeding state was also indicated 166 Population density. The rectums of bugs that had starved for 60 days (21 weeks pi) each contained about 86,900 T. cruzi (range 2,500 264,100) (Table 1), and the rectal lumen contained about 66,400 flagellates, i.e., about 75% of the total population. One day after feeding the bugs, the total rectal population had not changed. The next day the rectal population was reduced to about 51,100 flagellates/bug and on the third daf the number had decreased slightly again, ranging between 2,400 and 75,000. Within the first three daf, the number of flagellates in the lumen and on the wall was similar. fterwards, the population increased more than three times to about 101,200 flagellates/bug at five daf and to about 336,000 flagellates/bug at 10 daf. This increase was mainly due to the increase of the attached flagellates, about twothirds of the total population. Only the mean number of flagellates at 10 daf in the lumen and on the wall was significantly different from other populations (P 0.05, by uncan s test).
T. RUZI FTER FEEING OF STRVE VETORS 167 TLE 1 Mean S (range) (in thousands) of Trypanosoma cruzi hile 5 in different regions of the digestive tract of Triatoma infestans (fifth instar of 10 bugs) after a starvation period of 60 days (0), and 1, 2, 3, 5, and 10 days after feeding ays* Rectum Lumen Wall Total 0 66.4 62.6 (1.6 180.0) 20.5 26.0 (0.9 84.1) 1 42.4 39.9 49.0 37.2 (0.8 114.8) (0.6 120.0) 2 25.5 26.0 25.6 38.1 (0.3 86.6) (0.3 129.4) 3 14.8 13.0 18.0 18.6 (0.9 38.3) (1.5 57.0) 5 38.0 17.5 63.3 40.2 (15.1 68.8) (6.4 157.5) 10 104.4 56.2 231.7 217.9 (18.0 198.0) (31.7 754.0) * ays after feeding the bugs, which had starved for 60 days. 86.9 82.3 (2.5 264.1) 91.4 71.9 (1.4 228.2) 51.1 47.3 (0.6 157.2) 33.2 25.3 (2.4 75.0) 101.2 43.0 (58.6 184.5) 336.0 225.1 (49.7 849.7) evelopmental stages. When the percentages of the different developmental stages in the population of bugs that had starved for 60 days were compared, trypomastigotes dominated in the rectal lumen and on the wall with about 50%, followed by 20% epimastigotes (Table 2). Spheromastigotes were significantly more frequent in the rectal lumen (23%) than on the wall (7%) (P 0.001, by Student s t- test). rop-like intermediate stages were more common (11% and 15%) than slender intermediate stages (2%). One daf of starved bugs, the percentages of the different stages of T. cruzi in the rectal lumen and on the wall showed a statistically significant change (P 0.05, by uncan s test), with the exception of the slender intermediates (Table 3). Epimastigotes dominated, with 73% and 78%, respectively, of the total population of parasites. The percentage of trypomastigotes had decreased to less than 10%, and spheromastigotes and drop-like intermediate stages were less than 6% for both the lumen and the wall. stage previously not observed, giant cells, comprised 9% of the parasite population (range 1 19%). These giant cells possess a high cytoplasmic mass, many nuclei, kinetoplasts, and free flagella. etween one and two daf, the percentages of trypomastigotes, spheromastigotes, and drop-like intermediate forms showed only a small but statistically significant change (P 0.05, by uncan s test) within that period of time. The percentage of slender intermediates did not change in the lumen, but there was a statistically significant increase of parasites attached to the wall (P 0.001, by t-test). The percentage of giant cells showed a statistically significant increase (P 0.05, by uncan s test) to about 38% (range 10 70%) in the rectal lumen and 22% (range 14 30%) on the rectal wall. etween two and three daf, the percentages of the different developmental stages had not changed in the rectal lumen with the exception of the slender intermediates (Table 2). t three, five, and 10 daf, the percentages of epimastigotes and spheromastigotes did not change. The percentage of trypomastigotes, however, showed a statistically significant increase (P 0.05, by uncan s test) in the lumen from 5 15% to 35% and on the wall from 12 29% to 52%. The TLE 2 Percentage (mean S) of different developmental stages of Trypanosoma cruzi hile 5 in the rectal lumen and wall of Triatoma infestans (fifth instar) after a starvation period of 60 days (0), and 1, 2, 3, 5, and 10 days after feeding Rectal wall Rectal lumen ays after feeding* Trypomastigotes 47 10 9 5 4 4 5 2 15 5 35 14 56 21 6 3 8 6 12 3 29 11 52 11 Epimastigotes 18 8 73 7 53 16 64 8 65 4 57 13 20 14 78 8 60 2 38 13 57 15 41 10 Epi-trypomastigotes 2 1 5 3 3 4 0 0 0 0 4 3 2 1 1 1 6 1 1 0 2 1 3 2 Sphero-/epi-, or trypomastigotes 11 5 0 0 0 1 0 0 5 1 2 2 15 6 0 0 3 3 0 1 4 2 2 1 Spheromastigotes 23 5 2 2 1 1 0 0 2 2 2 1 7 4 6 6 0 1 2 2 2 2 2 1 Giant cells 0 0 9 4 38 20 30 7 13 8 0 0 0 0 9 6 22 7 47 4 6 4 0 0 N 5 5 6 3 5 4 5 5 4 6 6 9 * ays after feeding the bugs, which had starved for 60 days. N number of bugs examined.
168 KOLLIEN N SHU TLE 3 uncan s multiple range test comparison* of percentages of the respective developmental stages of Trypanosoma cruzi in the rectum of Triatoma infestans after a starvation period of 60 days (0), and 1, 2, 3, 5, and 10 days after feeding Trypomastigotes Epimastigotes Epi-trypomastigotes Sphero-/epi-, or trypomastigotes Spheromastigotes Giant cells ays E E * Geometric means of logarithm were compared after adding 0.25 to the 0 values. ata of groups with the same letters are not significantly different (P 0.05) in the respective developmental stages. percentage of giant cells showed a statistically significant decrease between three and five daf in the lumen and on the wall (P 0.05, by uncan s test) from 30% to 13% and from 47% to 6%, respectively, and disappeared 10 daf. ISUSSION When investigating parasite-invertebrate interactions between T. cruzi and the vector, the nutritional state of the bug is very important since it directly affects the flagellate that develops in the intestinal tract. 2, 10 12 The effects of starvation or feeding of the bug on T. cruzi, i.e., effects on the activity of the flagellate, the population density, and the population composition, have been recently summarized. 3, 10 The development of T. cruzi in the vector is affected by several factors, e.g., the strain of T. cruzi, the species and instar of bugs, the dose and duration of infection, and the temperature. 11, 13, 14 etailed investigations of a system in which the T. cruzi and the vector strain originate from the same locality have only been performed with our system. The bugs had been collected in 1979 and were separated initially into three different stocks. To avoid inbreeding effects, the three original unselected stocks were mixed starting in 1991. Since the susceptibility of the bugs could have changed during 15 years of breeding and by mixing the three stocks, it is important to compare the present results for agreement with those previously obtained. The effects of starvation and feeding on T. cruzi were previously investigated by Schaub in a long-term study after infection of the second instar of Triatoma infestans using a susceptible strain that had been selected from the three initial stocks and subsequent regular feeding conditions (intervals of 21 or 28 days). 2 fter starvation periods of 21 or 28 days, the population density of the flagellates in the small intestine was lower than seven days before. 2 fter a starvation period of 42 days (14 weeks pi), the rectums of fifth instars of Triatoma infestans, using one of the three initial stocks, were colonized by an average of 350,000 parasites. 12 fter the same starvation period, but 15 weeks pi, the progeny of a mixture of the three Triatoma infestans strains contained about 200,000 flagellates. In three groups of the same batch, which were fed or examined after in vitro diuresis, the average total rectal population per bug was 400,000, 350,000, and 300,000 flagellates, respectively. 10 fter starvation periods of 20, 30, and 60 daf (15, 16, and 20 weeks pi) one L5 of Triatoma infestans (mixed strain) from a similar batch contained an average of 300,000, 350,000, and 120,000 flagellates/rectum, respectively. These numbers were further decreased by increasing starvation periods. 3 In the present investigation, the conditions were identical to the previous starvation study and the numbers of flagellates found were similar. The distribution of the flagellates in the rectum of starved bugs was somewhat different in the previous investigations. t 28 and 42 daf, one-third of the T. cruzi population had been present in the rectal lumen of Triatoma infestans (L5). 2, 12 More recently, using the mixed strain of bugs, at 60 daf about two-thirds of the population was present in the rectal lumen, 3 and in the present investigation, after the same starvation period, the majority (75%) of the T. cruzi population was also in the rectal lumen. Short-term effects of feeding were obvious even 4 hr after feeding: the total rectal population of T. cruzi was reduced by more than 50% and only one-seventh of the parasites were present in the rectal lumen. 12 fter feeding, feces and urine were collected from individual bugs by restraining them above plastic tubes for a period of 3 hr. These excreta contained 30% of the original T. cruzi population; after feeding, the population in the lumen was about one-third of the total T. cruzi population. 10 In the earlier long-term investigation mentioned above, the effects of refeeding, in addition to those of starvation, were also evident. In L3 and L4, the number of parasites in the rectum increased within seven days after feeding or remained constant. The L5 showed a decrease at seven daf, but about one week later the number of parasites increased. 2 In the present investigation, feeding of the starved L5 resulted in a four-fold increase in the population density within 10 days. In addition to effects on the population density of T. cruzi, the distribution of the different developmental stages of the flagellate was also affected by starvation and feeding. In regularly fed bugs with an established infection, the population composition was different from that in starved bugs. 3, 10 In most bugs, the epimastigote form was dominant, or the numbers of epimastigotes and trypomastigotes were similar. Spheromastigotes and the two intermediate forms were rare. 2, 10 Starvation up to 60 days caused a dominance of trypomastigotes (50%), and epimastigotes (20%) were less common than under regular feeding conditions. The percentage of spheromastigotes had increased during the starvation period from less than 5% to approximately 20%. Each of the two intermediate stages (slender and drop-like) reached up to 10% in starved bugs. 3 In the present investigation, the population composition of the developmental stages of T. cruzi 60 days after feeding was similar to that found in the previous study, 3 and only the percentage of the slender intermediate stages (2%) was significantly lower. n investigation of the short-term effects of feeding bugs starved for 40 days, i.e., changes within 3 or 4 hr, showed that the percentages of trypomastigotes, spheromastigotes, and drop-like intermediates decreased, while the fractions of epimastigotes and slender intermediates increased, but no multiple divisions were recorded. 10, 12
T. RUZI FTER FEEING OF STRVE VETORS 169 FIGURE 1. evelopment of the different stages of Trypanosoma cruzi in the rectum of Triatoma infestans after starvation (dotted arrows) and a short (dashed arrow) or long (solid arrows) period of time after feeding the starved bugs. In the present investigation, feeding of bugs starved for 60 days resulted 24 hr later in an increase in the percentage of epimastigotes (70 80%) and the appearance of 1 6% spheromastigotes or of the two intermediate stages. The percentage of trypomastigotes decreased within the first three daf but then increased. Giant cells resembling a multiple division form appeared in high numbers for a short period of time. Since the development of a high number of giant cells in the population of T. cruzi was described in the present study for the first time, we want to focus the discussion on this developmental stage, pointing out where multiple divisions had been mentioned in earlier reports, at what time of the life-cycle of the parasite they appeared, and from which stage(s) they originated, and consider what their biological importance is. The giant cells have been mentioned in a few previous investigations (lvarenga NJ, Universidade Minas Gerais, elo Horizonte, razil, unpublished data), 15 but their proportion in the total rectal population and their localization within the intestinal tract have not been discussed. rack 15 found cells with 6 8 nucleoli and kinetoplasts 24 hr after infection, and multiple divisions of T. cruzi were also described by lvarenga (unpublished data) up to eight days pi. The present study showed that the development of giant cells of T. cruzi can be induced if long-term starved bugs were fed. In the initial development of T. cruzi after the infection of the vector, the multiple division stages developed in the small intestine. 15 lvarenga (unpublished data) also investigated the bugs a short time after the infection, but he did not separate the different parts of the intestinal tract. In the present investigation, the giant cells were found in the rectum of the bug. ue to the long starvation period, there were no parasites colonizing the small intestine. Two pathways in the development of giant cells have been proposed. 1) Spheromastigotes develop to stumpy epimastigotes, which then multiply either by equal divisions (resulting into two slender epimastigotes) or by a multiple division. The multiple divisions develop to rosettes, from which slender epimastigotes originate. 15 2) lvarenga (unpublished data) suggested that the multiple division stages of T. cruzi originate from spheromastigotes and should lead to epimastigotes, trypomastigotes, and drop-like stages. Since, in the present investigation, the percentage of spheromastigotes and drop-like intermediate stages decreased strongly within 24 hr after feeding, but no adequate increase in giant cells took place, the giant cells could not have originated from these round forms. However, the decrease in round forms within the first 24 hr seemed to be correlated with the increase of epimastigotes, and between 24 and 48 hr after feeding the percentages of epimastigotes decreased whereas those of giant cells increased. Since giant cells had been observed only rarely in previous investigations, the high percentage of these cells was surprising (up to 70% in individual bugs), and they appeared in high numbers for a short period of time after feeding. The development of giant cells in both previous investigations 15 (lvarenga NJ, unpublished data) seems to be due to the optimal feeding conditions after an infection, resulting in a rapid colonization. In the present investigation, an optimal food supply was present but after long starvation periods. The strong improvement in the nutritional conditions induced the generation of cytoplasm and cell organelles uncoupled from cell division processes. The following division restored the population, whose numbers had been reduced by the previous starvation. The present study shows for the first time that the development of a usually rare stage of T. cruzi, the giant cells, can be induced by feeding of starved vectors. The spheromastigotes and drop-like stages develop into epimastigotes, and the epimastigotes develop into these multiple division stages (Figure 1). Using biochemical and molecular biologic analyses, e.g., a detailed analyses of the amino acid and protein/peptide composition of the intestinal fluid and the genes of the digestive enzymes it is now possible to identify the inducing factors and the regulatory genes. cknowledgments: We appreciate many helpful comments on the English translation of the manuscript by r. R. assada, and the help with the statistics by r. G. Skipka. uthors address: strid H. Kollien and Günter. Schaub, epartment of Special Zoology, Ruhr University, -44780 ochum, Germany. REFERENES 1. hagas, 1909. Nova tripanozomiaze humana. Über eine neue Trypanosomiasis des Menschen. Mem Inst Oswaldo ruz 1: 159 218. 2. Schaub G, 1989. Trypanosoma cruzi: quantitative studies of development of two strains in the small intestine and rectum of the vector Triatoma infestans. Exp Parasitol 68: 260 273. 3. Kollien H, Schaub G, 1998. The development of Trypanosoma cruzi (Trypanosomatidae) in the reduviid bug Triatoma infestans (Insecta): influence of starvation. J Eukaryot Microbiol 45: 59 63. 4. Schaub G, Schottelius J, 1984. Identification of trypanosomes isolated from Reduviidae from North hile. Z Parasitenk 70: 3 9. 5. Schaub G, 1988. evelopmental time and mortality of larvae
170 KOLLIEN N SHU of Triatoma infestans infected with Trypanosoma cruzi. Trans R Soc Trop Med Hyg 82: 94 96. 6. öker, Schaub G, 1984. Scanning electron microscopic studies of Trypanosoma cruzi in the rectum of its vector Triatoma infestans. Z Parasitenk 70: 459 469. 7. Schaub G, 1990. The effect of lastocrithidia triatomae (Trypanosomatidae) on the reduviid bug Triatoma infestans: influence of group size. J Invertebr Pathol 56: 249 257. 8. Schaub G, 1983. lastocrithidia triatomae (Protozoa, Flagellata) - Triatoma infestans (Insecta, Reduviidae). Nachweis der Verschiedenen Entwicklungsstadien. öckeler W, Wülker W, eds. Parasitologisches Praktikum. Weinheim: Verlag hemie, 1 6. 9. Weber E, 1961. Grundriss der iologischen Statistik. Fourth edition. Jena: Fischer Verlag. 10. Kollien H, Schaub G, 1997. Trypanosoma cruzi in the rectum of the bug Triatoma infestans: effects of blood ingestion of the vector and artificial diuresis. Parasitol Res 83: 781 788. 11. sin S, atala S, 1995. evelopment of Trypanosoma cruzi in Triatoma infestans: influence of temperature and blood consumption. J Parasitol 81: 1 7. 12. Schaub G, Lösch P, 1988. Trypanosoma cruzi: origin of metacyclic trypomastigotes in the urine of the vector Triatoma infestans. Exp Parasitol 65: 174 186. 13. lvarenga NJ, rener Z, 1978. evelopment of Trypanosoma cruzi in the vector in the absence of blood. cta Trop 35: 315 317. 14. Garcia ES, zambuja P e, 1991. evelopment and interaction of Trypanosoma cruzi within the insect vector. Parasitol Today 7: 240 244. 15. rack, 1968. Elektronenmikroskopische Untersuchungen zum Lebenszyklus von Trypanosoma cruzi. cta Trop 25: 289 356.