JOURNAL
OF INVERTEBRATE
PATHOLOGY
56,249-257
(1990)
The Effect of Blastocrithidia triatomae (Trypanosomatidae) on the Reduviid Bug Triatoma infestam: influence of Group Size GONTER A. SCHAUB Institut fiir Biologie I (Zoologie), Albert-Ludwigs-Universitiit, Albertstrasse 21a, D-7800 Freiburg, Federal Republic of Germany Received October 17, 1989; accepted December 11, 1989 Developmental time and mortality in uninfected larvae of the reduviid bug Triatoma infestans and in those infected by feeding a mixture of blood and cysts of the homoxenous trypanosomatid Blastocrithidia triatomae were compared. Larvae were maintained isolated in 77-cm3 (area 9.6 cm*) beakers or in groups of 20,30,40, and 50 bugs per l-liter beaker (area 722 cm*). In uninfected groups, only a minor proportion of isolated bugs showed delayed development, but in groups infected with B. triatomae, additionally, a retardation in groups of 50 larvae occurred. Infected bugs needed more time to complete development in fourth and fifth instar than did uninfected bugs. Mean mortality rates of about 10% in uninfected groups were unaffected by group size. Mortality rates in most groups of infected bugs were about 50%, but in groups consisting of 50 larvae a statistically significant higher mortality rate of 75% was observed. This indicates a subpathological overcrowding stress, increased by the synergistic action of the flagellate. Q IWOAcademic RCSS, IOC. KEY WORDS: Triatoma infestans; Blastocrithidia triatomae; pathogenic trypanosomatid; larval development; larval mortality; overcrowding; isolation; Chagas’ disease.
INTRODUCTION
Triatoma infestans is the most important vector of Trypanosoma cruzi, the causative agent of Chagas’ disease in man (Zeledbn and Rabinovich, 1981). Like other important vectors, T. infestans is strongly affected by the homoxenous trypanosomatid Blastocrithidia triatomae which develops in the intestinal tract and Malpighian tubules of the hematophagous reduviid bugs (Schaub and Baker, 1986; Schaub, 1988a; Schaub and Breger, 1988; Schaub and Jensen, 1990). In infected adults, life span and reproduction rate are reduced (Schaub et al., 1984). In infected larvae, excretion, digestion, and tanning are disturbed; starvation resistance and encapsulation reaction of foreign material in the haemocoel are reduced; and developmental time and mortality rate are increased (Schaub and Schnitker, 1988; Schnitker et al., 1988; Schaub and Losch, 1989; G. A. Schaub, unpubl.) . The two last effects, in particular, are also known to occur in insect populations as a result of overcrowding stress (see re-
view by Solomon, 1957; Peters and Barbosa, 1977). In our series of studies concerned with the pathogenicity of B. triatomae for T. infestans, each experimental group consisted of about 40 bugs. Although the control groups were of the same size, an influence of group size was possible since overcrowding “is likely to cause the outbreak of disease” (Steinhaus, 1958a) and synergistic effects of parasite and crowding stress might occur. In such studies an influence of further factors like accessibility to sufficient food has to be carefully considered (Peters and Barbosa, 1977). MATERIALS
AND METHODS
The original strain of B. triatomae is maintained in T. infestuns or cultivated in vitro on a cell line of this bug (Reduth et al., 1989). Two passages after the initiation of an in vitro culture, cysts were harvested and cryopreserved at -80°C for l-2 months. The T. infestans strain “Chile” originates from Cachiyuyu, Chile, and was collected in 1979 (Schaub, 1988b). The bugs were fed on hens and reared at 26” f 1°C 249 0022-201 l/90 $1.50 Copyri&t 0 1990 by Academic Press, Inc. AII tights of reproduction in any form mscrvcd.
250
GijNTER
5040% relative humidity, and on a 16hr/&hr day/night rhythm. About 3-7 weeks after hatching, three large batches of first instars were fed on hens or on a mixture of B. triatomae and human blood (lo6 cyst stages/cm3 blood). (For convenience, the different stages are indicated as Ll , first larval stage, and so on until L5, the fifth larval stage.) One hundred twenty-five uninfected Ll and 125 infected Ll, all fed to capacity, were isolated singly into small beakers (height 8 cm, diameter 3.5 cm). The filter paper at the bottom had an area of 9.6 cm2. For calculation of mean developmental times or mortalities, these isolated control larvae and infected larvae were both divided into five groups, each consisting of 25 first instars. Other bugs were separated into groups consisting of different numbers of first instars. There were six uninfected groups, each with about 20 bugs, and five groups with about 30,40, or 50 bugs. With the exception of four groups with about 50 bugs, the same number of groups were obtained from the infected first instars. The groups were maintained in l-liter beakers, each containing a filter paper at the bottom and two crossed pieces of cardboard. The beakers were closed with nylon gauze. Thereby, the bugs could live on an area of 157 cm2 at the bottom and the top of the beaker and of 565 cm2 on the cardboard. Pieces of filter paper with feces from uninfected bugs were placed at the bottom of all beakers to allow an uptake of symbionts. Each group was blood fed weekly on hens and mice for l-2 hr, with the exception of the second, fifth, and eighth week, when there were bugs that had moulted within 4 days previously (Schaub and Breger, 1988). For feeding, the five groups of the 25 bugs maintained in isolation were each grouped in a one-liter beaker. Thus, these bugs had contact with other bugs for 1-2 hr/week. In infected groups, the first adults to moult and any dead bugs and in control groups the last adults to moult and any dead bugs were dissected for microscopical examination for B. triatomae (Schaub, 1988a).
A.
SCHAUB
Moults and mortality were recorded daily until adults emerged. From these data the times of occurrence of the first and last moults of each instar were calculated. As comparisons of such data can be complicated by a single occurrence of an extreme variation, the percentage of bugs in the respective instars was calculated for each group and at weekly intervals. The mean percentages (incl. SD) of uninfected and infected groups in the 2 weeks following the week in which moults first occurred are shown separately in Tables 2 and 3. For statistical comparisons of means of moults or mortality rates, Duncan’s multiple range test was calculated using the software of SAS Institute Inc. (Cary, North Carolina). RESULTS
B. triatomae developed in all bugs which had sucked the mixture of cysts and blood. No transmission to control groups occurred. Development Rates of Uninfected Larvae The first moults of each instar were observed 9 (Ll), 10 (L2), 12 (L3), 14 (L4), and 28 (LS) days after first feeding of the respective stage. The period between first and last moults increased for each successive instar (Table 1). The low standard deviations of the mean values indicate the homogeneous development within different groups of the same size. Group size did not influence the occurrence of first and last moults. (The highest values of 175 and 206 days originate from individual retarded bugs.) With the exception of L2 bugs kept singly, the means did not differ significantly (Duncan’s test; P < 0.05). A careful examination of the percentages of the different developmental stages at different weeks after first feeding showed that the development of an increasing percentage of isolated T. infestans larvae was retarded: Moulting occurred at least 1 week later in single insects than in groups consisting of 20, 30, 40, or 50 bugs which developed similarly (Tables 2, 3).
32-37 37.8 * 2.6A 3-9
54-55 57.2 f 4.448 61-81
7a-a5 87.0 -t 9O4ec 93-116
115-121 129.8 f 9.14 135-175
First L2 50% of L2
First L3 50% of L3
Fiit L4 50% of L4
First L5 50% of LS
114-120 124.5 k 4.14 13fb159
77-79 79.5 + 0.5CD 91-138
54-57 56.2 f 1.2*e 59-91
32-36 35.0 a 1.3BC 35-43
W-12 11.5 f 1.2e 11-17
4.tiBCD
93-109 114116 120.1 f l.lA 137-152
f
7a-al 83.4
53-55 54.8 k o.ae 58-79
33-35 35.0 + 0.7gc 35-53
9-15 13.2 It_ 2.94B 11-21
30
Uninfected
INSTARS OF
114-116 120.6 r 0.9 147-168
112-120 124.0 2 2.a* 14&2O6
76-79 78.6 + 1.F” 87-l 12
53-55 54.6 f 1.18 61-83
53-54 54.6 + 0.58 66-97 76-a2 al.2 + 3.6BCD -93-115
32-34 34.0 + 1.oc 40-50
lo-11 11.6 t 0.58 15-19
50
33-34 35.2 r 0.4BC 43-50
9-14 13.2 * 2.98 ii-28
40
123-127 146.2 k 6.O= 153-286
78-92 103.0 2 10.4 105-W
55-56 64.0 f 6.0 7O-102
35-37 39.6 f 1.9 4a-55
1 l-16 16.4 f 4.r 16-39
1 (25)
Timing of larval mouW
TABLE 1 Triatoma infestans IN GROUPS OF DIFFERENT triatomae INFECTION
121-138 141.5 + 7.3E 1 la-243
7MO 88.0 + 2.7AB 87-167
5455 56.2 f l.WB 6O-77
32-34 34.5 f 1.2=c 4Wi2
IO-13 13.7 T 2.1A 15-20
20
SIZE,
signiticantly different in Duncan’s test (P c 0.05) in the respective o Days after frst feeding of Ll. b Extreme values of the four or five groups. c Mean k SD, calculated from the days on which SO% of the larvae in the groups had moulted.
Note. Data of groups with same letters are not statistically
Last L5
Last LA
Last L3
Last L2
Last Lib
IO-17 15.6 + 4.44 13-25
of
20
FOR THE DIFFERENT
1 (25)
OF MOULTS
First Lib 50% of Ll’
Mocks
Size of groups:
PERIODS
30
Infected
instar.
12O-124 142.0 + 9.aB 155-260
7886 93.4 k 3.7A 1%180
53-55 56.2 + 1.3*e 66-a2
3635 36.4 2 0.54B 42-47
40
117-123 142.2 f 13.6B 219-247
78-85 93.2 2 7.SA 101-158
54-55 56.6 f 1.F 71-79
33-35 36.2 k O.aAB 43-48
50
115-143 152.8 f 16.2B 131-227
77-a2 92.8 i: a.zA 12O-231
S&55 58.5 -c 1.3* 55-74
32-33 34.3 f. 1.5c 4O-53
l&13 14.0 c_ z.oA,* 16-21
Blastocrithidia
lo-13 12.8 + t sAB ;5-22
AND WITHOUT
lc-14 13.4 ? 2.3AB 15-21
WITH
252
GiiNTER
PERCENTAGES
OF THE
DIFFERENT
WITHOUT Blastocrithidia
SCHAUB
TABLE 2 DEVELOPMENTAL STAGES~ OF ISOLATED Triatoma infestans. WITH triatomae INFECTION, AT DWFERENT WEEKS AFTER FIRST FEEDNG
Uninfected
Weeks after first feeding 3 4 6 7 9 10 13 14 18 19
A.
Infected
Ll
L2
L3
L4
LS
Adults
15 f 22 1*2 122 122 -
85 + 22 99r2 16222 225 2’4 122 122 152 -
-
-
-
-
83 2 22 9725 21 + 30 7*10 -
77 2 33 93211 27 + 29 19 f 25 2?2 l-+2
72 + 30 81?26 55 f 29 38 k 29
43 r 28 61 ? 30
d Mean f SD calculated from the uercentaaes of the five ’ First instar (Ll) to ftith instar (Lj) and a<s.
Development
AND
Ll
L2
L3
L4
LS
2122679’26 222 -
9822 34 2 26 9?6 122 12 2 -
66 f 26 9126 54231 28 + 22 3?5 3+4 l-t-2 -
45232 71 2 24 82~24 58+31 1329 13 k 9
15?22 39+30 83210 79 ” 10
Adults 4+2 8 f 4
croups.
Rates of Infected Larvae
If we first compare the data from infected groups, the times for those consisting of 20, 30,40, or 50 bugs were nearly identical, but development of many isolated larvae was retarded (Table 1). With the exception of Ll and L5 bugs, infected bugs kept isolated developed significantly more slowly than infected bugs kept in groups (Duncan’s test; P < 0.05). The percentages of different instars at different weeks after first feeding also showed the above-mentioned retarded development in isolated bugs (Tables 2, 3). Note that often four different instars were present at the same time in all infected groups. It was also clear that 35-40 weeks after the first feeding of Ll of groups consisting of 50 bugs, only about half of those infected bugs from which adults emerged had moulted (data not shown). In all other infected groups, 20% more bugs had emerged by that time. If we now compare data from uninfected and B. triutomae-infected groups, the times of the first and last moults of the first instars were nearly identical. However, the times of the first L5 moults and last L4 and L5 moults indicate retarded development of the infected groups (Table 1). Individual infected L5 often needed nearly double the time to complete larval development. Tables 2 and 3 demonstrate that development
of a high percentage retarded.
of infected bugs was
Mortality In the five to six uninfected groups, only one or two bugs sometimes died in the first four instars, but there were nearly always mortalities in the fifth instar. In three groups, 3,4, or 10 L5 died. (These data are included in the mean percentages in Table 4.) The total mortality rates of groups of different size did not differ significantly (Duncan’s test; P < 0.05) (Table 4). The infected groups showed only slightly higher mortality rates in the first three instars than did the uninfected groups (Table 4). Then, usually three to four died in each group at the L4 stage. Data for instarspecific mortality rates (I in Table 4) strongly indicate the effects of infection, especially in the last instar, in which about 50% of the infected larvae died. Mean total mortalities of infected groups differed significantly from the data of uninfected groups (Table 4). Only the mean total mortality of groups consisting of 50 bugs was significantly different from data from all other infected groups (Duncan’s test; P < 0.05). DISCUSSION
Development rate, size, fecundity, and mortality rate of insects are all known to be
INFLUENCE
* .
OF
I I I I I I Ii%
*-zoo---I +I +I +I +I +I +I tl xg2s-----
;I
tt
I I I I I I I I
t-mzw 1 1 1 1 (
1 tl +I +I +I 867W
P‘r4t-m-.-
(
1 1 1 +I tl +I +I +I +I gge,oo
N-P‘PI
I I +’ tl t’ +I I I I I ,gwm
;I 3
I I I I I I I I
Blastocrilhidia
ON
Triatoma
253
influenced by the size of groups used in the respective investigation (Chauvin, 1967; Peters and Barbosa, 1977). Besides overcrowding, maintenance in isolation can also affect them. The occurrence of such effects depends on the natural life style of the species or its developmental stages. Retarded larval development and high mortalities can be a result of grouping insects together that normally live singly, but they can also result from isolating or overcrowding insects which normally live gregariously (Peters and Barbosa, 1977). This effect of group size can change during development, e.g., in larvae of calliphorid flies which live gregariously during the feeding instars but singly in the third instar after the feeding phase when they crawl alone to pupation sites (Meyer, 1971). Since reduviid bugs belong to the group of gregarious insects, only data from this group will be considered in the following discussion of data from uninfected insects. Isolation causes an increase in mortality rate and/or a retardation of development in such different insects as Lepidoptera, crickets, cockroaches (literature summarized by Peters and Barbosa, 1977) and lygaeid bugs (Feir, 1963). A heavy mortality among isolated first instar larvae of Triatoma sanguisuga texana was noted by Pippin (1970) but no detailed data, especially concerning supply of symbionts, were given. We have now demonstrated for the first time this isolation effect on development of triatomines. The lack of effect on development of first instars described recently (Schaub, 1988~) was also confirmed in the present study. The apparent slight delay of first moults of isolated Ll seems to be a result of a slightly insufficient feeding. Those groups which moulted particularly late were started from a batch of Ll fed 7 weeks after eclosion. Although they seemed fully engorged after first feeding, many of them sucked once more in the following week. Recently I demonstrated that first instars of T. infestans do not engorge fully if they have starved for more than 6 weeks after eclosion (G. A. Schaub, un-
254
GiiNTER
A.
SCHAUB
NmN-wim~cc srnz +I
+I
tl
+I
tl
+I
tl
NNmmN-0\lr)~c+a +I +I +I
tl
fl
P(NmmPI-O\IO;~
+I
+I ptl 2 -+I
mN-Nc0O-rDP-O
+I
+I
tl -
+I m ;; m
tl -
tl v
+I mN'r4twlcvt.t-*~ +I +I
r4tmtmmu2t-~P~ +I +I +I
;;
+I Ici +I
-Plrcrwl-4*-P-000
+I
+I
fl -
$I m +I m +I
--a--PI--r.m
+I
+I
+I
tl
+I p +I
--a--rd.--r.ms +, +I
+I
tl
+I
NNr4NNNNN1r4+I tl
+I 0 +I
tl
tl
+I
tl
+I u-l tl
+I
+I
$I
tl
+I
+I P +I
tl
t,
+I
tl
+I \o tl
fl
+I
fl
m*NmOO--m*
+I
tl
mtNmOO--nYrf tl +I 00NmNm-Nmm
+I
+I
OONmNm-Nmw=+I tl +I -NNv-P(NNP‘h(
+I
+I
-NNt-Nc4NceN~ tl tl
INFLUENCE
OF
Blustocrithidia
publ). However, in addition to the initial delay of about 15%, which might be caused by insufficient feeding, further 5-15% moulted about 1 week later to L3, L4, L5, and adults and also in groups obtained from other batches of first instars. This indicates a slight isolation effect. Such a delay of a minor percentage of isolated bugs was also observed in a similar previous study (G. A. Schaub and S. Wolf, unpubl). Presumably, this influence would be more pronounced in bugs that get no contact with other specimens, even during feeding. I emphasize that isolation effects should be excluded in determinations of developmental times of triatomines. Just as does isolation, overcrowding exerts effects on the development of insects (Peters and Barbosa, 1977). Under these conditions, Drosophila needed more time to complete larval development and more specimens died (Scheiring et al., 1984; Moya and Castro, 1986). In overcrowded larvae of Calliphora, delayed metamorphosis correlates with a delayed release of ecdysone (Berreur et al., 1979). A more detailed study of the neuroendocrine mechanisms that delayed pupation of Drosophila, maintained under overcrowded or other adverse conditions, demonstrated a blocked secretion of the prothoracicotropic hormone, a reduced ecdysone secretion and activity of the peritracheal gland, and a decreased juvenile hormone content (Rauschenbach et al., 1987). In the reduviid bug Rhodnius prolixus, development in groups, which ranged from 2 to 128 larvae per 3.8-liter jar, has been studied by Rodriguez and Rabinovich (1980). They found a minor, though statistically significant, decrease of larval development time at the lowest and highest densities but no effect on mortality rates. This acceleration of development is in contrast to all studies previously discussed. In the present investigation with T. infestam, even densities of 50 bugs/liter did not influence development times and mortality rates of uninfected bugs. (Only when crowded bugs were also stressed by infection were
ON
Triatoma
255
there observable effects and even then only in groups of 50 but not of 40 bugs.) Crowded T. infestans and R. prolixus only showed an increase of developmental time and mortality rate if feeding time was limited (G6mez-Nufiez, 1964; Schofield, 1982). The correlation of outbreaks of insect diseases and crowding has been emphasized by Steinhaus (1958a, b, 1960). He mainly considered viral and bacterial diseases and he discussed the “mechanisms by which disease is made manifest in insects as the result of crowding” (Steinbaus, 1958a); he considered that there was an activation of the causative agent of disease, or an increase either in insect susceptibility, in the rate of transmission, or in concentrations of infectious material. Since we investigated orally infected bugs, none of these factors apply but a weakening of the insects has to be considered. Thus, the sensitivity to B. triatomae infections or the susceptibility to other pathogens which cannot develop in uninfected bugs or only do so at a harmless level might increase. The effects of keeping insects in isolation were not observed by Steinhaus (1958a) who investigated four species of insects but used populations consisting of infected and uninfected specimens. Three species seem to live singly and, therefore, the percentage of deaths increased with the number of infected larvae (1, 2, 5, 25, or 50 per container). Even two insects/container showed a higher mortality rate compared to those maintained in isolation. The presumably gregarious Californian oakworm reacted identically when 1, 2, 5, or 25 larvae were maintained per container, but mortality in groups of 50 was much higher. In our present study the influence of B. triutomae on development of L2, L3, and L4 is stronger in isolated bugs than in those in groups of 20,30, or 40, but the mortality rate is not higher in the crowded bugs. Overcrowding increases pathological effects in different infected insects. In addition to the virus- and bacteria-infected larvae of caterpillars and oakworms (Steinhaus, 1958a; Jaques, 1962), crowding
256
GiiNTER
increased the mortality rate of a predatory mite which had been infected with the weak bacterial pathogen Serratia marcescens (Lighthart et al., 1988). An increase of mortality rates of infected larvae also occurred in Microsporidia-infected larvae of a corn borer (Siegel et al., 1986). In the present study, a slightly synergistic action of overcrowding and infection with the homoxenous trypanosomatid B. triatomae was found in developmental times of larvae maintained in groups of 50. In three of four such groups, first moults to L5 occurred about l-3 weeks later than in the other infected groups. A strong synergistic action was evident in mortality rates in groups of 50 larvae which were significantly increased. We cannot conclusively exclude additional stress factors, but comparison of our data with results from other groups (summarized in Schaub, 1988a) show that we had an optimal development and extraordinary low mortality rates in our control groups. Therefore, the additional increase of mortality rates of B. triatomaeinfected larvae in the overcrowded groups does demonstrate the importance of group size. Our data indicate that isolation and overcrowding are important stress factors for bugs. They are less important in natural populations of bugs but overcrowding especially should be avoided in insectaria rearing bugs for xenodiagnosis. Both stress factors have to be considered in investigations of the pathological effects of pathogens on reduviid bugs. In our previous investigations (see Introduction), groups consisted of about 40 larvae/group. Since this group size showed no synergistic effects with B. triatomae, pathological effects should occur similarly in natural populations. The more sensitive reaction of infected bugs to the stress of overcrowding implies that other stress factors may well also act synergistically with B. triatomae. In natural populations, adverse temperatures or relative humidities might affect B. triatomaeinfected bugs more strongly. Therefore, B.
A. SCHAUB
triatomae should be tested as a potential biological control for important vectors of Chagas’ disease. ACKNOWLEDGMENTS The author thanks Mrs. E. Walter for typing the manuscript, Mr. A. Hag for his help in statistical analysis, and Mrs. A. Hoffmann for her accurate technical assistance. I am most grateful to Dr. S. Maddrell, Cambridge, for many helpful comments on the English translation. The support of the Deutsche Forschungsgemeinschaft (project No. DFG Scha 339/l-3) is gratefully acknowledged.
REFERENCES BERRELJR, P., PORCHERON, P., BERREURBONNENFANT, J., AND DRAY, F. 1979. External factors and ecdysone release in Calliphoru eryfhrocephala. Experienta, 35, 1031. CHAUVIN, R. 1967. “Die Welt der Insekten.” Kindler, Milnchen. FEIR, D. 1963. Effects of rearing alone and in groups on the growth of the milkweed bug, Oncopeltus jiisciatus (Hemiptera: Lygaeidea). Ann. Entomol. Sot. Amer., 56, 406-407. GOMEZ-Nu~~Ez, J. C. 1964. Mass rearing of Rhodnius prolixus. Bull. WHO, 31, 565-567. JAQUES, R. P. 1%2. Stress and nuclear polyhedrosis in crowded populations of Trichoplusia ni (Hiibner). J. Insect Pathol., 4, l-22. LIGHTHART, B., SEWALL, D., AND THOMAS, D. R. 1988. Effect of several stress factors on the susceptibility of the predatory mite, Metaseiulus occident&is (Atari: Phytoseiidae), to the weak bacterial pathogen Serratia marcescens. J. Znvertebr. Pathol., 52, 3342. MEYER, S. G. E. 1971. “Der respiratorische Stoffwechsel von Callitroga macellaria (Fabricius), (Diptera, Calliphoridae), in Abhtigigkeit von Entwickung und Umwelt.” Ph.D. thesis, Universitit Bonn. MOYA, A., AND CASTRO, J. A. 1986. Larval competition in Drosophila melanogaster: The model of the bands of density. Oikos, 47, 280-286. PETERS, T. M., AND BARBOSA, P. 1977. Influence of population density on size, fecundity, and developmental rate of insects in culture. Annu. Rev. Entomol., 22, 431450. PIPPIN, W. F. 1970. The biology and vector capability of Triatoma sanguisuga texana Usinger and Triatoma gerstaeckeri (St& compared with Rhodnius prolirus (Stti) (Hemiptera: Triatominae). J. Med. Entomol., I, 30-45. RAUSCHENBACH, I. Y., LUKASHINA, N. S., MAKSIMOVSKY, L. F., AND KOROCHKIN, L. I. 1987. Stress-like reaction of Drosophila to adverse environmental factors. J. Comp. Physiol. B, 157, 519531.
INFLUENCE
Blastocrithidia ON Triatoma
D., SCHAUB, G. A., AND F%JDNEY, M. 1989. Cultivation of Blastocrithidia triatomae (Trypanosomatidae) on a cell line of its host Triatoma infestans (Reduviidae). Parasitology, 98, 387-393. RODRIGUEZ, D., AND RABINOVICH, J. 1980. The effect of density on some population parameters of Rhodnius prolixus (Hemiptera: Reduviidae) under laboratory conditions. J. Med. Entomol., 17, 165-171. SCHAUB, G. A. 1988a. Developmental time and mortality in larvae of the reduviid bugs Triatoma infestans and Rhodnius prolixus after coprophagic infection with Blastocrithidia triatomae (Trypanosomatidae). J. Znvertebr. Pathol., 51, 23-31. SCHAUB, G. A. 1988b. Direct transmission of Trypanosoma cruzi between vectors of Chagas’ disease. Acta Trop., 45, 11-19. SCHAUB, G. A. 1988~. Development of isolated and group-reared first instars of Triatoma infestans infected with Trypanosoma cruzi. Parasitol. Res., 74, 593-594. SCHAUB, G. A., AND BIKER, C. A. 1986. Scanning electron microscopic studies of Blastocrithidia triatomae (Trypanosomatidae) in the rectum of Triatoma infestans (Reduviidae). J. Protozool., 33,266 270. SCHAUB, G. A., AND BREGER, B. 1988. Pathological effects of Blastocrithidia triatomae (Trypanosomatidae) on the reduviid bugs Triatoma sordida, T. pallidipennis and Dipetalogaster maxima after COprophagic infection. Med. Vet. Entomol., 2, 309318. SCHAUB, G. A., AND JENSEN, C. 1990. Developmental time and mortality of the reduviid bug Triatoma infestans with differential exposure to coprophagic infections with Blastocrithidia triatomae (Trypanosomatidae). J. Znvertebr. Pathol., 55, 17-27. SCHAUB, G. A., AND LOSCH, P. 1989. Parasite/ host-interrelationships of the trypanosomatids Trypanosoma cruzi and Blastocrithidia triatomae and the reduviid bug Triatoma infestans: Influence of starvation of the bug. Ann. Trop. Med. Parasitol., 83, 215-223. SCHAUB, G. A., MEISER, A., AND WOLF, S. 1984. The REDUTH,
‘\,
OF
257
influence of Blastocrithidia triatomae (Trypanosomatidae) on reproduction of Triatoma infestans (Reduviidae). In “Abstracts XVII. Int. Congr. Entomol., Hamburg 1984,” p. 675. SCHAUB, G. A., AND SCHNITKER, A. 1988. Intluence of Blastocrithidia triatomae (Trypanosomatidae) on the reduviid bug Triatoma infestans: Alterations in the Malpighian tubules. Parasitol. Res., 75, 8W7. SCHEIRING, J. F., DAVIS, D. G., RANASINGHE, A., AND TEA& C. A. 1984. Effects of larval crowding on life history parameters in Drosophila melanogaster Meigen (Diptera: Drosophilidae). Ann. Entomol. Sot. Amer., 77, 329-332. SCHNITKER, A., SCHAUB, G. A., AND MADDRELL, S. H. P. 1988. The influence of Blastocrithidia triatomae (Trypanosomatidae) on the reduviid bug Triatoma infestans: In vivo and in vitro diuresis and production of diuretic hormone. Parasitology, %, 9-17. SCHOFIELD, C. J. 1982. The role of blood intake in density regulation of populations of Triatoma infestans (Klug) (Hemiptera: Reduviidae). Bull. Entomol. Res., 72, 617-629. SIEGEL, J. P., MADDOX, J. V., AND RUESINK, W. G. 1986. Lethal and sublethal effects of Nosema pyrausta on the European corn borer (Ostrinia nubilalis) in central Illinois. J. Znvertebr. Pathol., 48, 167-173. SOLOMON, M. E. 1957. Dynamics of insect populations. Annu. Rev. Entomol., 2, 121-142. STEINHAUS, E. A. 1958a. Crowding as a possible stress factor in insect disease. Ecology, 39.503-514. STEINHAUS, E. A. 1958b. Stress as a factor in insect disease. Proc. Tenth Znt. Conf Entomol., 4, 725730. STEINHAUS, E. A. 1960. Symposium: Selected topics in microbial ecology. II. The importance of environmental factors in the insect-microbe ecosystem. Bacterial. Rev., 24, 365-313. ZELEDON, R., AND RABINOVICH, J. E. 1981. Chagas’ disease: An ecological appraisal with special emphasis on its insect vectors. Annu. Rev. Entomol., 26, 101-133.
OF INVERTEBRATE
PATHOLOGY
56,249-257
(1990)
The Effect of Blastocrithidia triatomae (Trypanosomatidae) on the Reduviid Bug Triatoma infestam: influence of Group Size GONTER A. SCHAUB Institut fiir Biologie I (Zoologie), Albert-Ludwigs-Universitiit, Albertstrasse 21a, D-7800 Freiburg, Federal Republic of Germany Received October 17, 1989; accepted December 11, 1989 Developmental time and mortality in uninfected larvae of the reduviid bug Triatoma infestans and in those infected by feeding a mixture of blood and cysts of the homoxenous trypanosomatid Blastocrithidia triatomae were compared. Larvae were maintained isolated in 77-cm3 (area 9.6 cm*) beakers or in groups of 20,30,40, and 50 bugs per l-liter beaker (area 722 cm*). In uninfected groups, only a minor proportion of isolated bugs showed delayed development, but in groups infected with B. triatomae, additionally, a retardation in groups of 50 larvae occurred. Infected bugs needed more time to complete development in fourth and fifth instar than did uninfected bugs. Mean mortality rates of about 10% in uninfected groups were unaffected by group size. Mortality rates in most groups of infected bugs were about 50%, but in groups consisting of 50 larvae a statistically significant higher mortality rate of 75% was observed. This indicates a subpathological overcrowding stress, increased by the synergistic action of the flagellate. Q IWOAcademic RCSS, IOC. KEY WORDS: Triatoma infestans; Blastocrithidia triatomae; pathogenic trypanosomatid; larval development; larval mortality; overcrowding; isolation; Chagas’ disease.
INTRODUCTION
Triatoma infestans is the most important vector of Trypanosoma cruzi, the causative agent of Chagas’ disease in man (Zeledbn and Rabinovich, 1981). Like other important vectors, T. infestans is strongly affected by the homoxenous trypanosomatid Blastocrithidia triatomae which develops in the intestinal tract and Malpighian tubules of the hematophagous reduviid bugs (Schaub and Baker, 1986; Schaub, 1988a; Schaub and Breger, 1988; Schaub and Jensen, 1990). In infected adults, life span and reproduction rate are reduced (Schaub et al., 1984). In infected larvae, excretion, digestion, and tanning are disturbed; starvation resistance and encapsulation reaction of foreign material in the haemocoel are reduced; and developmental time and mortality rate are increased (Schaub and Schnitker, 1988; Schnitker et al., 1988; Schaub and Losch, 1989; G. A. Schaub, unpubl.) . The two last effects, in particular, are also known to occur in insect populations as a result of overcrowding stress (see re-
view by Solomon, 1957; Peters and Barbosa, 1977). In our series of studies concerned with the pathogenicity of B. triatomae for T. infestans, each experimental group consisted of about 40 bugs. Although the control groups were of the same size, an influence of group size was possible since overcrowding “is likely to cause the outbreak of disease” (Steinhaus, 1958a) and synergistic effects of parasite and crowding stress might occur. In such studies an influence of further factors like accessibility to sufficient food has to be carefully considered (Peters and Barbosa, 1977). MATERIALS
AND METHODS
The original strain of B. triatomae is maintained in T. infestuns or cultivated in vitro on a cell line of this bug (Reduth et al., 1989). Two passages after the initiation of an in vitro culture, cysts were harvested and cryopreserved at -80°C for l-2 months. The T. infestans strain “Chile” originates from Cachiyuyu, Chile, and was collected in 1979 (Schaub, 1988b). The bugs were fed on hens and reared at 26” f 1°C 249 0022-201 l/90 $1.50 Copyri&t 0 1990 by Academic Press, Inc. AII tights of reproduction in any form mscrvcd.
250
GijNTER
5040% relative humidity, and on a 16hr/&hr day/night rhythm. About 3-7 weeks after hatching, three large batches of first instars were fed on hens or on a mixture of B. triatomae and human blood (lo6 cyst stages/cm3 blood). (For convenience, the different stages are indicated as Ll , first larval stage, and so on until L5, the fifth larval stage.) One hundred twenty-five uninfected Ll and 125 infected Ll, all fed to capacity, were isolated singly into small beakers (height 8 cm, diameter 3.5 cm). The filter paper at the bottom had an area of 9.6 cm2. For calculation of mean developmental times or mortalities, these isolated control larvae and infected larvae were both divided into five groups, each consisting of 25 first instars. Other bugs were separated into groups consisting of different numbers of first instars. There were six uninfected groups, each with about 20 bugs, and five groups with about 30,40, or 50 bugs. With the exception of four groups with about 50 bugs, the same number of groups were obtained from the infected first instars. The groups were maintained in l-liter beakers, each containing a filter paper at the bottom and two crossed pieces of cardboard. The beakers were closed with nylon gauze. Thereby, the bugs could live on an area of 157 cm2 at the bottom and the top of the beaker and of 565 cm2 on the cardboard. Pieces of filter paper with feces from uninfected bugs were placed at the bottom of all beakers to allow an uptake of symbionts. Each group was blood fed weekly on hens and mice for l-2 hr, with the exception of the second, fifth, and eighth week, when there were bugs that had moulted within 4 days previously (Schaub and Breger, 1988). For feeding, the five groups of the 25 bugs maintained in isolation were each grouped in a one-liter beaker. Thus, these bugs had contact with other bugs for 1-2 hr/week. In infected groups, the first adults to moult and any dead bugs and in control groups the last adults to moult and any dead bugs were dissected for microscopical examination for B. triatomae (Schaub, 1988a).
A.
SCHAUB
Moults and mortality were recorded daily until adults emerged. From these data the times of occurrence of the first and last moults of each instar were calculated. As comparisons of such data can be complicated by a single occurrence of an extreme variation, the percentage of bugs in the respective instars was calculated for each group and at weekly intervals. The mean percentages (incl. SD) of uninfected and infected groups in the 2 weeks following the week in which moults first occurred are shown separately in Tables 2 and 3. For statistical comparisons of means of moults or mortality rates, Duncan’s multiple range test was calculated using the software of SAS Institute Inc. (Cary, North Carolina). RESULTS
B. triatomae developed in all bugs which had sucked the mixture of cysts and blood. No transmission to control groups occurred. Development Rates of Uninfected Larvae The first moults of each instar were observed 9 (Ll), 10 (L2), 12 (L3), 14 (L4), and 28 (LS) days after first feeding of the respective stage. The period between first and last moults increased for each successive instar (Table 1). The low standard deviations of the mean values indicate the homogeneous development within different groups of the same size. Group size did not influence the occurrence of first and last moults. (The highest values of 175 and 206 days originate from individual retarded bugs.) With the exception of L2 bugs kept singly, the means did not differ significantly (Duncan’s test; P < 0.05). A careful examination of the percentages of the different developmental stages at different weeks after first feeding showed that the development of an increasing percentage of isolated T. infestans larvae was retarded: Moulting occurred at least 1 week later in single insects than in groups consisting of 20, 30, 40, or 50 bugs which developed similarly (Tables 2, 3).
32-37 37.8 * 2.6A 3-9
54-55 57.2 f 4.448 61-81
7a-a5 87.0 -t 9O4ec 93-116
115-121 129.8 f 9.14 135-175
First L2 50% of L2
First L3 50% of L3
Fiit L4 50% of L4
First L5 50% of LS
114-120 124.5 k 4.14 13fb159
77-79 79.5 + 0.5CD 91-138
54-57 56.2 f 1.2*e 59-91
32-36 35.0 a 1.3BC 35-43
W-12 11.5 f 1.2e 11-17
4.tiBCD
93-109 114116 120.1 f l.lA 137-152
f
7a-al 83.4
53-55 54.8 k o.ae 58-79
33-35 35.0 + 0.7gc 35-53
9-15 13.2 It_ 2.94B 11-21
30
Uninfected
INSTARS OF
114-116 120.6 r 0.9 147-168
112-120 124.0 2 2.a* 14&2O6
76-79 78.6 + 1.F” 87-l 12
53-55 54.6 f 1.18 61-83
53-54 54.6 + 0.58 66-97 76-a2 al.2 + 3.6BCD -93-115
32-34 34.0 + 1.oc 40-50
lo-11 11.6 t 0.58 15-19
50
33-34 35.2 r 0.4BC 43-50
9-14 13.2 * 2.98 ii-28
40
123-127 146.2 k 6.O= 153-286
78-92 103.0 2 10.4 105-W
55-56 64.0 f 6.0 7O-102
35-37 39.6 f 1.9 4a-55
1 l-16 16.4 f 4.r 16-39
1 (25)
Timing of larval mouW
TABLE 1 Triatoma infestans IN GROUPS OF DIFFERENT triatomae INFECTION
121-138 141.5 + 7.3E 1 la-243
7MO 88.0 + 2.7AB 87-167
5455 56.2 f l.WB 6O-77
32-34 34.5 f 1.2=c 4Wi2
IO-13 13.7 T 2.1A 15-20
20
SIZE,
signiticantly different in Duncan’s test (P c 0.05) in the respective o Days after frst feeding of Ll. b Extreme values of the four or five groups. c Mean k SD, calculated from the days on which SO% of the larvae in the groups had moulted.
Note. Data of groups with same letters are not statistically
Last L5
Last LA
Last L3
Last L2
Last Lib
IO-17 15.6 + 4.44 13-25
of
20
FOR THE DIFFERENT
1 (25)
OF MOULTS
First Lib 50% of Ll’
Mocks
Size of groups:
PERIODS
30
Infected
instar.
12O-124 142.0 + 9.aB 155-260
7886 93.4 k 3.7A 1%180
53-55 56.2 + 1.3*e 66-a2
3635 36.4 2 0.54B 42-47
40
117-123 142.2 f 13.6B 219-247
78-85 93.2 2 7.SA 101-158
54-55 56.6 f 1.F 71-79
33-35 36.2 k O.aAB 43-48
50
115-143 152.8 f 16.2B 131-227
77-a2 92.8 i: a.zA 12O-231
S&55 58.5 -c 1.3* 55-74
32-33 34.3 f. 1.5c 4O-53
l&13 14.0 c_ z.oA,* 16-21
Blastocrithidia
lo-13 12.8 + t sAB ;5-22
AND WITHOUT
lc-14 13.4 ? 2.3AB 15-21
WITH
252
GiiNTER
PERCENTAGES
OF THE
DIFFERENT
WITHOUT Blastocrithidia
SCHAUB
TABLE 2 DEVELOPMENTAL STAGES~ OF ISOLATED Triatoma infestans. WITH triatomae INFECTION, AT DWFERENT WEEKS AFTER FIRST FEEDNG
Uninfected
Weeks after first feeding 3 4 6 7 9 10 13 14 18 19
A.
Infected
Ll
L2
L3
L4
LS
Adults
15 f 22 1*2 122 122 -
85 + 22 99r2 16222 225 2’4 122 122 152 -
-
-
-
-
83 2 22 9725 21 + 30 7*10 -
77 2 33 93211 27 + 29 19 f 25 2?2 l-+2
72 + 30 81?26 55 f 29 38 k 29
43 r 28 61 ? 30
d Mean f SD calculated from the uercentaaes of the five ’ First instar (Ll) to ftith instar (Lj) and a<s.
Development
AND
Ll
L2
L3
L4
LS
2122679’26 222 -
9822 34 2 26 9?6 122 12 2 -
66 f 26 9126 54231 28 + 22 3?5 3+4 l-t-2 -
45232 71 2 24 82~24 58+31 1329 13 k 9
15?22 39+30 83210 79 ” 10
Adults 4+2 8 f 4
croups.
Rates of Infected Larvae
If we first compare the data from infected groups, the times for those consisting of 20, 30,40, or 50 bugs were nearly identical, but development of many isolated larvae was retarded (Table 1). With the exception of Ll and L5 bugs, infected bugs kept isolated developed significantly more slowly than infected bugs kept in groups (Duncan’s test; P < 0.05). The percentages of different instars at different weeks after first feeding also showed the above-mentioned retarded development in isolated bugs (Tables 2, 3). Note that often four different instars were present at the same time in all infected groups. It was also clear that 35-40 weeks after the first feeding of Ll of groups consisting of 50 bugs, only about half of those infected bugs from which adults emerged had moulted (data not shown). In all other infected groups, 20% more bugs had emerged by that time. If we now compare data from uninfected and B. triutomae-infected groups, the times of the first and last moults of the first instars were nearly identical. However, the times of the first L5 moults and last L4 and L5 moults indicate retarded development of the infected groups (Table 1). Individual infected L5 often needed nearly double the time to complete larval development. Tables 2 and 3 demonstrate that development
of a high percentage retarded.
of infected bugs was
Mortality In the five to six uninfected groups, only one or two bugs sometimes died in the first four instars, but there were nearly always mortalities in the fifth instar. In three groups, 3,4, or 10 L5 died. (These data are included in the mean percentages in Table 4.) The total mortality rates of groups of different size did not differ significantly (Duncan’s test; P < 0.05) (Table 4). The infected groups showed only slightly higher mortality rates in the first three instars than did the uninfected groups (Table 4). Then, usually three to four died in each group at the L4 stage. Data for instarspecific mortality rates (I in Table 4) strongly indicate the effects of infection, especially in the last instar, in which about 50% of the infected larvae died. Mean total mortalities of infected groups differed significantly from the data of uninfected groups (Table 4). Only the mean total mortality of groups consisting of 50 bugs was significantly different from data from all other infected groups (Duncan’s test; P < 0.05). DISCUSSION
Development rate, size, fecundity, and mortality rate of insects are all known to be
INFLUENCE
* .
OF
I I I I I I Ii%
*-zoo---I +I +I +I +I +I +I tl xg2s-----
;I
tt
I I I I I I I I
t-mzw 1 1 1 1 (
1 tl +I +I +I 867W
P‘r4t-m-.-
(
1 1 1 +I tl +I +I +I +I gge,oo
N-P‘PI
I I +’ tl t’ +I I I I I ,gwm
;I 3
I I I I I I I I
Blastocrilhidia
ON
Triatoma
253
influenced by the size of groups used in the respective investigation (Chauvin, 1967; Peters and Barbosa, 1977). Besides overcrowding, maintenance in isolation can also affect them. The occurrence of such effects depends on the natural life style of the species or its developmental stages. Retarded larval development and high mortalities can be a result of grouping insects together that normally live singly, but they can also result from isolating or overcrowding insects which normally live gregariously (Peters and Barbosa, 1977). This effect of group size can change during development, e.g., in larvae of calliphorid flies which live gregariously during the feeding instars but singly in the third instar after the feeding phase when they crawl alone to pupation sites (Meyer, 1971). Since reduviid bugs belong to the group of gregarious insects, only data from this group will be considered in the following discussion of data from uninfected insects. Isolation causes an increase in mortality rate and/or a retardation of development in such different insects as Lepidoptera, crickets, cockroaches (literature summarized by Peters and Barbosa, 1977) and lygaeid bugs (Feir, 1963). A heavy mortality among isolated first instar larvae of Triatoma sanguisuga texana was noted by Pippin (1970) but no detailed data, especially concerning supply of symbionts, were given. We have now demonstrated for the first time this isolation effect on development of triatomines. The lack of effect on development of first instars described recently (Schaub, 1988~) was also confirmed in the present study. The apparent slight delay of first moults of isolated Ll seems to be a result of a slightly insufficient feeding. Those groups which moulted particularly late were started from a batch of Ll fed 7 weeks after eclosion. Although they seemed fully engorged after first feeding, many of them sucked once more in the following week. Recently I demonstrated that first instars of T. infestans do not engorge fully if they have starved for more than 6 weeks after eclosion (G. A. Schaub, un-
254
GiiNTER
A.
SCHAUB
NmN-wim~cc srnz +I
+I
tl
+I
tl
+I
tl
NNmmN-0\lr)~c+a +I +I +I
tl
fl
P(NmmPI-O\IO;~
+I
+I ptl 2 -+I
mN-Nc0O-rDP-O
+I
+I
tl -
+I m ;; m
tl -
tl v
+I mN'r4twlcvt.t-*~ +I +I
r4tmtmmu2t-~P~ +I +I +I
;;
+I Ici +I
-Plrcrwl-4*-P-000
+I
+I
fl -
$I m +I m +I
--a--PI--r.m
+I
+I
+I
tl
+I p +I
--a--rd.--r.ms +, +I
+I
tl
+I
NNr4NNNNN1r4+I tl
+I 0 +I
tl
tl
+I
tl
+I u-l tl
+I
+I
$I
tl
+I
+I P +I
tl
t,
+I
tl
+I \o tl
fl
+I
fl
m*NmOO--m*
+I
tl
mtNmOO--nYrf tl +I 00NmNm-Nmm
+I
+I
OONmNm-Nmw=+I tl +I -NNv-P(NNP‘h(
+I
+I
-NNt-Nc4NceN~ tl tl
INFLUENCE
OF
Blustocrithidia
publ). However, in addition to the initial delay of about 15%, which might be caused by insufficient feeding, further 5-15% moulted about 1 week later to L3, L4, L5, and adults and also in groups obtained from other batches of first instars. This indicates a slight isolation effect. Such a delay of a minor percentage of isolated bugs was also observed in a similar previous study (G. A. Schaub and S. Wolf, unpubl). Presumably, this influence would be more pronounced in bugs that get no contact with other specimens, even during feeding. I emphasize that isolation effects should be excluded in determinations of developmental times of triatomines. Just as does isolation, overcrowding exerts effects on the development of insects (Peters and Barbosa, 1977). Under these conditions, Drosophila needed more time to complete larval development and more specimens died (Scheiring et al., 1984; Moya and Castro, 1986). In overcrowded larvae of Calliphora, delayed metamorphosis correlates with a delayed release of ecdysone (Berreur et al., 1979). A more detailed study of the neuroendocrine mechanisms that delayed pupation of Drosophila, maintained under overcrowded or other adverse conditions, demonstrated a blocked secretion of the prothoracicotropic hormone, a reduced ecdysone secretion and activity of the peritracheal gland, and a decreased juvenile hormone content (Rauschenbach et al., 1987). In the reduviid bug Rhodnius prolixus, development in groups, which ranged from 2 to 128 larvae per 3.8-liter jar, has been studied by Rodriguez and Rabinovich (1980). They found a minor, though statistically significant, decrease of larval development time at the lowest and highest densities but no effect on mortality rates. This acceleration of development is in contrast to all studies previously discussed. In the present investigation with T. infestam, even densities of 50 bugs/liter did not influence development times and mortality rates of uninfected bugs. (Only when crowded bugs were also stressed by infection were
ON
Triatoma
255
there observable effects and even then only in groups of 50 but not of 40 bugs.) Crowded T. infestans and R. prolixus only showed an increase of developmental time and mortality rate if feeding time was limited (G6mez-Nufiez, 1964; Schofield, 1982). The correlation of outbreaks of insect diseases and crowding has been emphasized by Steinhaus (1958a, b, 1960). He mainly considered viral and bacterial diseases and he discussed the “mechanisms by which disease is made manifest in insects as the result of crowding” (Steinbaus, 1958a); he considered that there was an activation of the causative agent of disease, or an increase either in insect susceptibility, in the rate of transmission, or in concentrations of infectious material. Since we investigated orally infected bugs, none of these factors apply but a weakening of the insects has to be considered. Thus, the sensitivity to B. triatomae infections or the susceptibility to other pathogens which cannot develop in uninfected bugs or only do so at a harmless level might increase. The effects of keeping insects in isolation were not observed by Steinhaus (1958a) who investigated four species of insects but used populations consisting of infected and uninfected specimens. Three species seem to live singly and, therefore, the percentage of deaths increased with the number of infected larvae (1, 2, 5, 25, or 50 per container). Even two insects/container showed a higher mortality rate compared to those maintained in isolation. The presumably gregarious Californian oakworm reacted identically when 1, 2, 5, or 25 larvae were maintained per container, but mortality in groups of 50 was much higher. In our present study the influence of B. triutomae on development of L2, L3, and L4 is stronger in isolated bugs than in those in groups of 20,30, or 40, but the mortality rate is not higher in the crowded bugs. Overcrowding increases pathological effects in different infected insects. In addition to the virus- and bacteria-infected larvae of caterpillars and oakworms (Steinhaus, 1958a; Jaques, 1962), crowding
256
GiiNTER
increased the mortality rate of a predatory mite which had been infected with the weak bacterial pathogen Serratia marcescens (Lighthart et al., 1988). An increase of mortality rates of infected larvae also occurred in Microsporidia-infected larvae of a corn borer (Siegel et al., 1986). In the present study, a slightly synergistic action of overcrowding and infection with the homoxenous trypanosomatid B. triatomae was found in developmental times of larvae maintained in groups of 50. In three of four such groups, first moults to L5 occurred about l-3 weeks later than in the other infected groups. A strong synergistic action was evident in mortality rates in groups of 50 larvae which were significantly increased. We cannot conclusively exclude additional stress factors, but comparison of our data with results from other groups (summarized in Schaub, 1988a) show that we had an optimal development and extraordinary low mortality rates in our control groups. Therefore, the additional increase of mortality rates of B. triatomaeinfected larvae in the overcrowded groups does demonstrate the importance of group size. Our data indicate that isolation and overcrowding are important stress factors for bugs. They are less important in natural populations of bugs but overcrowding especially should be avoided in insectaria rearing bugs for xenodiagnosis. Both stress factors have to be considered in investigations of the pathological effects of pathogens on reduviid bugs. In our previous investigations (see Introduction), groups consisted of about 40 larvae/group. Since this group size showed no synergistic effects with B. triatomae, pathological effects should occur similarly in natural populations. The more sensitive reaction of infected bugs to the stress of overcrowding implies that other stress factors may well also act synergistically with B. triatomae. In natural populations, adverse temperatures or relative humidities might affect B. triatomaeinfected bugs more strongly. Therefore, B.
A. SCHAUB
triatomae should be tested as a potential biological control for important vectors of Chagas’ disease. ACKNOWLEDGMENTS The author thanks Mrs. E. Walter for typing the manuscript, Mr. A. Hag for his help in statistical analysis, and Mrs. A. Hoffmann for her accurate technical assistance. I am most grateful to Dr. S. Maddrell, Cambridge, for many helpful comments on the English translation. The support of the Deutsche Forschungsgemeinschaft (project No. DFG Scha 339/l-3) is gratefully acknowledged.
REFERENCES BERRELJR, P., PORCHERON, P., BERREURBONNENFANT, J., AND DRAY, F. 1979. External factors and ecdysone release in Calliphoru eryfhrocephala. Experienta, 35, 1031. CHAUVIN, R. 1967. “Die Welt der Insekten.” Kindler, Milnchen. FEIR, D. 1963. Effects of rearing alone and in groups on the growth of the milkweed bug, Oncopeltus jiisciatus (Hemiptera: Lygaeidea). Ann. Entomol. Sot. Amer., 56, 406-407. GOMEZ-Nu~~Ez, J. C. 1964. Mass rearing of Rhodnius prolixus. Bull. WHO, 31, 565-567. JAQUES, R. P. 1%2. Stress and nuclear polyhedrosis in crowded populations of Trichoplusia ni (Hiibner). J. Insect Pathol., 4, l-22. LIGHTHART, B., SEWALL, D., AND THOMAS, D. R. 1988. Effect of several stress factors on the susceptibility of the predatory mite, Metaseiulus occident&is (Atari: Phytoseiidae), to the weak bacterial pathogen Serratia marcescens. J. Znvertebr. Pathol., 52, 3342. MEYER, S. G. E. 1971. “Der respiratorische Stoffwechsel von Callitroga macellaria (Fabricius), (Diptera, Calliphoridae), in Abhtigigkeit von Entwickung und Umwelt.” Ph.D. thesis, Universitit Bonn. MOYA, A., AND CASTRO, J. A. 1986. Larval competition in Drosophila melanogaster: The model of the bands of density. Oikos, 47, 280-286. PETERS, T. M., AND BARBOSA, P. 1977. Influence of population density on size, fecundity, and developmental rate of insects in culture. Annu. Rev. Entomol., 22, 431450. PIPPIN, W. F. 1970. The biology and vector capability of Triatoma sanguisuga texana Usinger and Triatoma gerstaeckeri (St& compared with Rhodnius prolirus (Stti) (Hemiptera: Triatominae). J. Med. Entomol., I, 30-45. RAUSCHENBACH, I. Y., LUKASHINA, N. S., MAKSIMOVSKY, L. F., AND KOROCHKIN, L. I. 1987. Stress-like reaction of Drosophila to adverse environmental factors. J. Comp. Physiol. B, 157, 519531.
INFLUENCE
Blastocrithidia ON Triatoma
D., SCHAUB, G. A., AND F%JDNEY, M. 1989. Cultivation of Blastocrithidia triatomae (Trypanosomatidae) on a cell line of its host Triatoma infestans (Reduviidae). Parasitology, 98, 387-393. RODRIGUEZ, D., AND RABINOVICH, J. 1980. The effect of density on some population parameters of Rhodnius prolixus (Hemiptera: Reduviidae) under laboratory conditions. J. Med. Entomol., 17, 165-171. SCHAUB, G. A. 1988a. Developmental time and mortality in larvae of the reduviid bugs Triatoma infestans and Rhodnius prolixus after coprophagic infection with Blastocrithidia triatomae (Trypanosomatidae). J. Znvertebr. Pathol., 51, 23-31. SCHAUB, G. A. 1988b. Direct transmission of Trypanosoma cruzi between vectors of Chagas’ disease. Acta Trop., 45, 11-19. SCHAUB, G. A. 1988~. Development of isolated and group-reared first instars of Triatoma infestans infected with Trypanosoma cruzi. Parasitol. Res., 74, 593-594. SCHAUB, G. A., AND BIKER, C. A. 1986. Scanning electron microscopic studies of Blastocrithidia triatomae (Trypanosomatidae) in the rectum of Triatoma infestans (Reduviidae). J. Protozool., 33,266 270. SCHAUB, G. A., AND BREGER, B. 1988. Pathological effects of Blastocrithidia triatomae (Trypanosomatidae) on the reduviid bugs Triatoma sordida, T. pallidipennis and Dipetalogaster maxima after COprophagic infection. Med. Vet. Entomol., 2, 309318. SCHAUB, G. A., AND JENSEN, C. 1990. Developmental time and mortality of the reduviid bug Triatoma infestans with differential exposure to coprophagic infections with Blastocrithidia triatomae (Trypanosomatidae). J. Znvertebr. Pathol., 55, 17-27. SCHAUB, G. A., AND LOSCH, P. 1989. Parasite/ host-interrelationships of the trypanosomatids Trypanosoma cruzi and Blastocrithidia triatomae and the reduviid bug Triatoma infestans: Influence of starvation of the bug. Ann. Trop. Med. Parasitol., 83, 215-223. SCHAUB, G. A., MEISER, A., AND WOLF, S. 1984. The REDUTH,
‘\,
OF
257
influence of Blastocrithidia triatomae (Trypanosomatidae) on reproduction of Triatoma infestans (Reduviidae). In “Abstracts XVII. Int. Congr. Entomol., Hamburg 1984,” p. 675. SCHAUB, G. A., AND SCHNITKER, A. 1988. Intluence of Blastocrithidia triatomae (Trypanosomatidae) on the reduviid bug Triatoma infestans: Alterations in the Malpighian tubules. Parasitol. Res., 75, 8W7. SCHEIRING, J. F., DAVIS, D. G., RANASINGHE, A., AND TEA& C. A. 1984. Effects of larval crowding on life history parameters in Drosophila melanogaster Meigen (Diptera: Drosophilidae). Ann. Entomol. Sot. Amer., 77, 329-332. SCHNITKER, A., SCHAUB, G. A., AND MADDRELL, S. H. P. 1988. The influence of Blastocrithidia triatomae (Trypanosomatidae) on the reduviid bug Triatoma infestans: In vivo and in vitro diuresis and production of diuretic hormone. Parasitology, %, 9-17. SCHOFIELD, C. J. 1982. The role of blood intake in density regulation of populations of Triatoma infestans (Klug) (Hemiptera: Reduviidae). Bull. Entomol. Res., 72, 617-629. SIEGEL, J. P., MADDOX, J. V., AND RUESINK, W. G. 1986. Lethal and sublethal effects of Nosema pyrausta on the European corn borer (Ostrinia nubilalis) in central Illinois. J. Znvertebr. Pathol., 48, 167-173. SOLOMON, M. E. 1957. Dynamics of insect populations. Annu. Rev. Entomol., 2, 121-142. STEINHAUS, E. A. 1958a. Crowding as a possible stress factor in insect disease. Ecology, 39.503-514. STEINHAUS, E. A. 1958b. Stress as a factor in insect disease. Proc. Tenth Znt. Conf Entomol., 4, 725730. STEINHAUS, E. A. 1960. Symposium: Selected topics in microbial ecology. II. The importance of environmental factors in the insect-microbe ecosystem. Bacterial. Rev., 24, 365-313. ZELEDON, R., AND RABINOVICH, J. E. 1981. Chagas’ disease: An ecological appraisal with special emphasis on its insect vectors. Annu. Rev. Entomol., 26, 101-133.
