SUSCEPTIBILITY OF ADULT HELIGMOSOMOIDES POLYGYRUS TO INTESTINAL INFLAMMATORY RESPONSES INDUCED BY HETEROLOGOUS INFECTION

Ah&raCt-BEHNKE J. M., CABAJ W. and WAKELIN D. 1992. Susceptibility of adult Heligmosomoides polygyrus to intestinal inflammatory responses induced by heterologous infection. International Journalfor Parasitology 22: 75-86. Adult H. polygyrus are capable of surviving for many months after primary exposure of mice to infective larvae, raising the possibility that worms of this species have inherent resistance to intestinal immune responses. Accordingly experiments were carried out to determine whether H. polygyrus are resistant to the inflammatory changes elicited during the acute phase of the intestinal response to Trichinellu spiralis. Adult worms were expelled from mice when their presence coincided with the most intense phase of inflammation elicited by T. spiralis. The effect was dose-dependent with more intense T. spiralis challenge resulting in a correspondingly greater loss of H. polygyrus. Even the less pathogenic species T. pseudospiralis elicited a response of sufficient intensity in NIH mice to cause the expulsion of H. polygyrus from concurrently infected animals. Tissue larval stages of H. polygyrus were protected from expulsion by their location deep in the intestinal walls and the maximum detrimental effect against H. polygyrus was observed during the adult phase or during the establishment of L3 larvae. Acceleration of the response to T. spiralis in immune challenged mice resulted in earlier loss of H. polygyrus. When the expulsion of T. spiralis was delayed (e.g. from slow responder C57BL/lO mice) the loss of H. polygyrus took place correspondingly later. These experiments demonstrate unequivocally that mouse strains which normally tolerate chronic infections with H.polygyrus have the capacity to mount intestinal inflammatory responses of sufficient vigour to remove the worms but that this potential is not normally realized. However, the observation that some H. polygyrus always survived even when the response induced by T. spiralis was of the rapid secondary type suggests that the parasites are resilient in the face of the inflammatory response capable of removing most of the worms. It is suggested that in addition to the immunomodulatory strategy employed by adult worms to prevent the intestinal response being elicited, the worms have a second line of defence which is reflected in their resilience to responses which they have been unable to prevent.


INTRODUCTION
THE intestinal inflammatory response elicited by Nippostrongylus brasiliensis and Trichinella spiralis has been the focus of attention for well over 4 decades and the immunological control of this response is well documented (Rothwell, 1989;Moqbel & MacDonald, 1990). It is established that the induction of the response is specific, requiring the sensitization of CD4+ T, lymphocytes (Crook, 1990;Grencis, Riedlinger & Wakelin, 1985), however, the final effecters remain as controversial as ever. Among recent additions to the list of possible effecters are leukotrienes (Douch, t To whom all correspondence should be addressed. Harrison, Buchanan & Greer, 1983), platelet activating factor (Moqbel & MacDonald, 1990), tumour necrosis factor (Ovington, 1987), mast cell protease II (McKean & Pritchard, 1989) and oxygen radicals (Smith & Bryant, 1989a,b). Once induced the specifically triggered effecters act non-specifically with detrimental consequences to unrelated parasites residing in the intestine at the time of expression (Christensen, Nansen, Fagbemi & Monrad, 1987). During the response to T. spiralis, totally unrelated nematodes (Nippostrongylus brasiliensis, see Kennedy, 1980) and cestodes (Hymenolepis diminuta, see Behnke, Bland & Wakelin, 1977;Christie, Wakelin & Wilson, 1979;H. nana Howard, Christie, Wakelin, Wilson & Behnke, 1978), may have their growth severely impaired. None of these show specific immunological cross immunity with T. spiralis and their loss from the host is brought about solely through the non-specific effecters of the inflammatory response initiated by T. spiralis and the resultant changes in the intestinal environment.
Not all nematodes parasitizing the intestines of vertebrates elicit acute inflammatory responses of the sort associated with the rejection of T. spiralis and N. brasiliensis. Some cause long-term chronic infections in which there is little/no evidence of hostprotective immunity. For example, Trichostrongylus tenuis accumulates in wild grouse and survives for over 600 days (Shaw & Moss, 1989), Necator americanus lives for up to 17 years in man (Palmer, 1955;Beaver, 1988) and Heligmosomoides polygyrus (Nematospiroides dubius) causes infections which last for 8-10 months in mice (Robinson, Wahid, Behnke & Gilbert, 1989). Little is known about the survival strategies of the former species, but H. polygyrus has been intensively investigated and it is generally accepted that this species is immunomodulatory, downregulating intestinal immunity to enable its own survival (Behnke, 1987). One of the indicators of intestinal inflammation, the mast cell response, is hardly detectable in mice carrying H. polygyrus and even in mice concurrently infected with T. spiralis mastocytosis is greatly depressed (Dehlawi, Wakelin & Behnke, 1987). This in turn is associated with prolonged infection with T. spiralis (Behnke, Wakelin & Wilson, 1978) indicating that H. polygyrus not only downregulates the inflammatory response in relation to its own survival but that the mechanism it invokes acts non-specifically and that it can benefit unrelated species resident in concurrently infected animals (Behnke, 1987;Christensen et al., 1987).
The effect of H. polygyrus in prolonging concurrent infections with heterologous species of parasites is well documented (Behnke, 1987;Christensen et al., 1987;Cabaj, 1989). However, the effect of an intense inflammatory response on H. polygyrus has never been studied in detail, although the worms are known to be resistant to free oxygen radical damage in vitro (Smith & Bryant, 1986). Furthermore, in comparison with nematodes causing acute infections, H. polygyrus have relatively high levels of oxygen radical scavenging enzymes and it has been suggested that in consequence the worms are more resilient to host-generated free oxygen radical-mediated damage, which may be a component of the non-specifically acting effecters of intestinal expulsion (Smith & Bryant, 1986). Thus in addition to showing immunomodulatory properties H. polygyrus may have a second line of defence to safeguard their survival in the host.
In the present paper we report the results of experiments which extend our earlier studies but this time examine the effect of the intestinal inflammatory response elicited by T. spiralis on concurrent infection with H.polygyrus. T.pseudospiralis, which is reputedly less pathogenic and causes longer lasting infections in mice than T. spiralis but also elicits an acute intestinal response (Przyjalkowski, Starzynski, Pykalo & Cabaj, 1981;Przyjalkowski & Pykalo, 1988;Stewart, Wood & Boley, 1985;Stewart, Mann, Ubelaker, McCarthy & Wood, 1988), was exploited as a further source of intestinal inflammation. The experiments were designed to minimize the possible immunodepressive influence of H. polygyrus, by keeping infections with the latter species low, worm burdens being just sufficient to allow any reductions to be recognized and analysed statistically, but not high enough to cause a major downregulation of the intestinal response to Trichinella spp. In contrast Trichinella infections were kept high to ensure maximum establishment and the induction of the host protective intestinal response. It is shown conclusively that adult H. polygyrus are lost when the parasites are caught in an inflammatory response which they have been unable to prevent. However, the same experiments again emphasize the tremendous resilience of this parasite, since expulsion of H. polygyrus was never complete and often only minimal, supporting the existence of a second line of defence against host immunity, as suggested by Smith & Bryant (1986).

MATERIALS AND METHODS
Mice. The mice used in this study were purchased from Harlan Olac Ltd and were bred in the departmental animal house. Unless otherwise stated, male mice were used throughout. All animals were housed under conventional animal house conditions with access to food and water ad libitum.
Parasite. T. spiralis was originally obtained in 1975 from the Wellcome Research Laboratories (Beckenham, Kent) since when it has been passaged regularly in CFLP mice. Our stain corresponds to H. polygyrus bakeri as reported by Durette-Desset, Kinsella & Forester (1972). The methods used for maintenance and infection have been reported previously (Jenkins & Behnke, 1977). Adult parasites were recovered by a 6 h incubation at 37'C of small intestines, suspended in a gauze, in 50 ml beakers containing Hanks's saline as described. Faecal egg counts were carried out by the method of Behnke & Parish (1979). The strain of T. spiralis and the methods used for the infection of mice and recovery of worms have been described previously (Wakelin & Lloyd, 1976). T. pseudospirah was obtained from the Polish Academy of Sciences and was maintained identically to T. spiralis.
Statistical analysis of results. The results are presented as group mean values (MWR) f standard error (s.E.M.). Nonparametric statistical procedures were used to analyse the data sets, because of small sizes (Sokal& Rohlf, 1969). When more than two groups required comparison at a single time point the Kruskal-Wallis statistic H was calculated to determine whether there was a significant treatment effect. If significant, specific groups were compared to the control group (or as stated) by the Mann-Whitney U test. Correlations between variables were tested by the Spearman Rank Order Correlation test and the statistic r, is given, as appropriate. Probabilities were calculated from statistics tables and are presented as follows: * P = 0.05; l *0.05 > P ao.02; ***0.02 > P 20.01; **** 0.01 > P > 0.001; l **** P <O.OOl. imposed on an existing adult population of H. polygyrus in the concurrently infected groups. It can be seen from the data which are summarized in Fig. 1 that the majority of T. spiralis were expelled at exactly the same time by mice carrying single and concurrent infections, although worms persisted longer in concurrently infected mice with 2.8 f 3.8 being recovered on day 21 compared with none in the single infection group (T. spiralis only). It is also evident from Fig. 1 that whilst H. polygyrus persisted without loss to day 21 post-challenge in mice infected only with this species, there was a considerable loss of worms from concurrently infected animals with an 82.9 and 95.6% reduction by days 11 and 2 1, respectively.

Relationship between intensity of T. spiralis (Expts 2-5) and T. pseudospiralis (Expts 6-7) infections and the proportion of H. polygyrus lost during intestinal inflammation
Four separate experiments were carried out to ascertain the importance of infection intensity with T. spiralis on the proportion of adult H. polygyrus lost during the ensuing inflammatory response in concurrently infected mice. All the experiments were carried out in NIH mice, each included control groups to monitor the establishment of T. spiralis in single and concurrent infection groups. In each, H. polygyrus was given 14 days before T. spiralis and the mice were killed 3 weeks after the latter infection. The results are summarized in Fig. 2A, in which the mean worm burdens recovered at autopsy are expressed as a percentage of the single infection control group (H. polygyrus only). Rank order correlation coefficients were also calculated and these are presented in Table 1. declined. The reduction in eggs per gram of faeces corresponded to the onset of the inflammatory response to Trichinella.

Consequence of varying the interval between and order of infection with H. polygyrus andT. spiralis (Expts 8-11)
All of the experiments described so far were based on a primary infection with H. polygyrus followed 14 days later by challenge with T. spiralis or T. pseudospiralis. The inflammatory response generated by Trichinella spp. will therefore have coincided in all cases with the presence of adult H. polygyrus. It was necessary to extend these observations to encompass the effect of inflammation on the larval stages of H. polygyrus. Four experiments were carried out in which the sequence of primary/challenge infection and the interval between primary and challenge were varied. The results from three such experiments are summarized in Fig. 4.
The trend is very similar in all three experiments, with almost identical results for two (Expts 8 and 9) and a slightly earlier sequence of events but a similarly shaped graph for Expt 10. The most impressive effect was always when H. polygyrus was administered well before T. spiralis such that the inflammatory response coincided with the presence of adult worms. As the interval between primary infection with H. polygyrus and challenge with T. spiralis was reduced, so the proportion of H. polygyrus surviving increased. When Three experiments were carried out in male NIH mice; 250 (Expts 9 and 10) or 400 muscle larvae (Expts 8) of T. spiralis were administered on day 0. Groups of five to six mice were infected with 100 L3 of H. polygyrus before, on the same day or after T. spiralis on the days shown and on each occasion a group of naive control mice was also given H. polygyrus. All the mice were killed for worm counts on day 28 post-T. spiralis infection.
Statistical analysis: MWRs from concurrently infected groups were compared with their respective single infection control group by the Mann-Whitney U test and points marked* were not significantly different from control groups (P > 0.05).
H. polygyrus was given 5 days after T. spiralis (Expts 8 and 9) or on the day of T. spirafis infection (Expt 10) relatively few worms were lost. However, as the interval between a primary T. spiraks infection and a challenge with H. polygyrus increased, so the proportion of worms surviving again decreased. Finally when the interval was 13 days (Expt 8), there was again relatively little loss of H. polygyrus.
A fourth experiment (Expt 11) examined the effect of administering the two species at the same time (day 0) and this confirmed the data from Expts 8 and 9, indicating that only 26% of the worms survived to day 21 (MWR control group = 114.8f 13.4, n = 5; MWR concurrently infected group = 29.8 f 9.3, n = 5). The secondary response to T. spiralis occurs earlier in immune NIH mice and is associated with more intense inflammation of the intestine than takes place during the primary response. If the principal reason for loss of H.polygyrus in concurrently infected mice is through the non-specifically acting mediators of the inflammatory response specifically elicited by T. spiralis, secondary responses to T. spiralis occurring in concurrently infected mice would be expected to exert a significantly more potent effect against H. polygyrus. This prediction was tested in three experiments.
All were carried out in male NIH mice. The basic design involved four groups of mice to control for possible specific as well as non-specific interactions. Thus each experiment had one group of mice which received only H. polygyrus (day -14) and therefore controlled for infectivity of this species (infectivity control). A second group received a primary infection with T. spiralis (days -34 or -35) and a second with H. polygyrus (day -14) to control for possible cross immunity between the species as well as residual activity persisting after expulsion of the primary T. spiralis infection (cross immunity control). A third group received a primary infection with H. polygyrus (day -14) followed by challenge with T. spiralis (day 0) as in earlier experiments.
The important experimental group comprised mice which received a primary infection with T. spiralis (day -34 or -35), followed by a second infection with H. polygyrus (day -14) and a third infection with T. spiralis (day 0). Additional control groups monitored the infectivity of T. spiralis in both infections and the expression of acquired resistance in the absence of concurrent infection.
The results from Expt 12 are summarized in Table 2. It is quite clear that there was no residual activity persisting from the primary T. spiralis and no cross immunity between the species, since the establishment of H. polygyrus was not impaired when T. spiralis was given 3 weeks earlier. However, both of the H. polygyrus infected groups challenged with T. spiralis on day 0 had significantly fewer H. polygyrus on day 7 and the MWR was lower in mice which had been primed with T. spiralis. As can be seen, control groups indicated that the primary T. spiralis infection had sensitized the mice and an accelerated expulsion of T. spiralis ensued when both single-species and concurrently infected mice were subjected to homologous challenge.
A second experiment (Expt 13) also monitored changes in both species on two occasions following challenge with T. spiralis ( H. polygyrus) 80.3 &6.6 (n=6). The T. spiralis infection given on day 0 comprised 250 muscle larvae of which 144.5 f 4.1 established and on this occasion, by day 5, there was no effect on H. polygyrus in mice which had not been primed with T. spiralis (MWR = 83.3 f 2.7, n = 6) although a small reduction was evident by day 28 (MWR = 72.3 f 6.7, not significant). However, mice primed with T. spiralis on day -35, challenged with H. polygyrus on day -14 and then with T. spiralis on day 0 only had 57.5 f 9.8 H. polygyrus on day 5 and 43.8 f 7.9 on day 28 (P = 0.008).

f 4.3 worms (n = 6) and on day + 28 (42 days after infection with
These two experiments were repeated on a larger scale with groups of mice from all the above treatments killed at regular intervals throughout the secondary infection with T. spiralis. The results are shown in Fig. 5. The data confirm and extend both of the above experiments and show that secondary responses to T. spiralis which take place in concurrently infected mice exert a rapid and potent effect against H. polygyrus causing earlier loss ofworms than when a primary T. spiralis infection is superimposed on existing adult H. polygyrus.

Comparison of the effect on H. polygyrus of primary responses to T. spiralis infast andslow responder mouse strains (Expts 15-l 7)
The final series of experiments exploited the known difference in the time of onset and subsequent intensity of the intestinal inflammatory response to T. spiralis between fast (NIH) and slow (C57BL/lO) responder strains (Wakelin, 1978), in a further attempt to link the loss of H. polygyrus from concurrently infected mice to the acute inflammatory response elicited by T. spiralis. It would be anticipated that concurrently infected NIH mice would lose a greater proportion of H. polygyrus and the loss would occur earlier than in C57BL/lO mice.
The predictions of the hypothesis were tested in three experiments, two of which are summarized in Table 3 (Expts 15 and 16). In Expt 15, a single infection dose of T. spiralis was used throughout, whereas in Expt 16, groups of concurrently infected mice were challenged with one of two doses as shown. There was also a difference in establishment as shown by control groups infected only with T. spiralis and killed before expulsion. Thus in Expt 15 establishment of T. spiralis was relatively poor, but despite this a significant reduction in the H. polygyrus worm burden was observed by day 28 post-challenge with T. spiralis in concurrently infected mice of both strains. However, proportionally the loss was marginally greater in NIH relative to C57BL/lOmice (33.9% vs 28.3%). When T. spiralis infections were more intense (Expt 16), significant loss of H. polygyrus was detected as early as 11 days post-challenge confirming earlier data (Fig. 1). Proportionally the loss was again greater in NIH mice.  By day 30 the differences between the two strains were even further exaggerated with a 93 and 79.3% reduction in H. polygyrus in NIH mice compared with the singly infected control group, but a loss of only 50.6 and 54.4% in C57BL/lO mice. These findings were confirmed in a larger experiment which followed the time course of events more precisely in both mouse strains and the results are summarized in Figs. 6 and 7. Again it is quite evident that H. polygyrus burdens already began to decline by day 10 in the fast responder NIH mice, at a time when T. spiralis had not yet been lost (Fig. 6~). Then, as T. spiralis was expelled, so a major proportion of H. polygyrus was also lost. In C57BL/lO mice, inflammation generated by T. spiralis was slower to appear and correspondingly loss of both parasites in concurrently infected animals and T. spiralis in single infection control groups was delayed. As in the earlier experiments (Expts 15 and 16) the loss from NIH mice was proportionally greater (day 20, NIH = 79.5%, C57BL/lO = 21.3%). The difference in the onset and intensity ofintestinal inflammation between the strains was also reflected in the effects it had on the fecundity of H. polygyrus in concurrently infected mice. Figure 7 shows that faecal egg counts declined in concurrently infected NIH mice as early as day 5 after challenge with T. spiralis whereas in C57BL/lO mice no appreciable reduction was evident before day 14 postchallenge.

The effect ofH. polygyrus on the expulsion ofr. spiralis from concurrently infected mice
Although the primary purpose of this study was to examine the effect of the intestinal inflammation elicited by T. spiralis on the survival of H. polygyrus in concurrent infections, control groups were included in all experiments to monitor the infectivity of T. spiralis and to confirm that expulsion took place at the expected times. In concurrently infected mice, both H. polygyrus and T. spiralis were counted at autopsy, enabling a comparison of T. spiralis MWRs between single and concurrent infection groups. In general the experiments were designed so as to minimize the influence of the known immunomodulatory effects of H. ~ofygyr~s (Behnke et al., 1978) and therefore, H. polygyrus infections were kept low (10&150 larvae in most experiments, although some utilized higher doses, e.g. Expt 1). Only some of the data are shown. Thus Fig. 1 shows that during Expt 1, there was no major effect on the expulsion of T. spiralis, although a few larvae did persist longer in the concurrent infection group. However, Figs. 5 and 6 show quite clearly that T. spiralis did persist longer in concurrently infected animals despite the low doses of H. polygyrus which were used on those occasions. In most of the remaining experiments for which we do not present quantitative data, T. spiralis and T. pseudospiralis persisted longer in the concurrent infection groups compared with mice infected only with Trichittella spp.

DISCUSSION
Despite the inability of most mouse strains to expel primary worm burdens of H. poiygyrus as rapidly as T. spiralis, the experiments reported in this paper clearly establish that two such strains (NTH and C57BL/lO) have the capacity to mount an intestinal i~ammat~ry response of sufficient intensity to cause the expulsion of many, in some experiments the majority, of HT. pofygyrus. Thus the important conclusion from our experiments is that these mouse strains have the components necessary to mount an intestinal inflammatory response capable of driving the adult worms out but normally fail to express the response and in consequence tolerate chronic primary infections with H. polygyrus. This conclusion again focuses attention on the importance of the immunomodulatory strategy utilized by adult H. pofygyrus in avoiding host immunity and it emphasizes that this strategy is essential to the parasite because H. palygyrus is susceptible to the non-specifically acting effecters of inflammatory responses which it has failed to prevent.
Our experiments defined the conditions under which expulsion of H. p~lygyr~ could occur. The intensity of the concurrent T. spiralis infection was an important factor, because although some loss occurred when a low intensity T. spiralis infection was used, this was not always predictable and consistent effects were observed only when the higher intensity infections were employed. There was a significant negative correlation between the proportion of H. polygyrus surviving the inflammatory phase of the T. spiralis challenge and the dose of T. spiralis administered, although this relationship was not linear as seen in Fig. 2. We explored also the effect of T. pseudospiralis which is reputed to be less pathogenic than T. spiralis (Przyjalkowski et al., 1981;Przyjalkowski & Pykalo, 1988) and to persist for longer after primary infection. However, NIH mice expel both species at about the same time and our experiments suggest that the inflammatory effect generated by both Trichineh spp. is comparably detrimental to the survival of H. polygyrus in concurrently infected animals.
One of the interesting observations was the variation in the proportion of H. polygyrus eliminated when the relative timing of infection with the two species was varied. The proportion of H. polygyrus expelled declined as larvae were given nearer to the time of infection with T. spiralis. This reduction can be explained by the position of its larval stages. When mice were infected with H. polygyrus on days O-5, L3 would have penetrated deeply into their normal site of development in the muscularis mucosa, and would have been relatively secure from expulsion at the time of peak inflammation, i.e. days 5-10, emerging from this site only after their normal development had been completed, some 8-10 days later (Bryant, 1973;Sukhdeo, O'Grady & Hsu, 1984). By this time conditions in the gut lumen would have returned almost to normal. The subsequent decline in survival when H. polygyrus larvae were given after T. spiralis on days 5-10 is explained by a failure to establish, since in these animals H. polygyrus L3 larvae would have been attempting to exsheath and penetrate the intestinal mucosa at the peak of the inflammatory phase. Finally the second decline in the proportion of H. pofygyrus being eliminated, occurring when larvae were given on days 11-15, is explained by the less inflamed environment following the expulsion of T. spiralis.
The final two series of experiments exploited situations in which the inflammatory response to T. spiralis was known to be altered in time. Thus mice experiencing a secondary T. spiralis infection mount a vigorous rapid inflammatory response (Wakelin & Lloyd, 1976;Alizadeh & Wakelin, 1982). In contrast poor responder mice such as C57BL/lO reject T. spiralis more slowly, later and with less vigour than fast responders such as NIH (Wakelin, 1980). The observations that adult H. polygyrus were lost earlier in NIH mice experiencing a secondary response to T. spiralis and considerably later in C57BL/lO mice are in accordance with our hypothesis that it is the non-specific effecters of the inflammatory response which make life untenable for H. polygyrus.
One more approach could have been used to confirm the requirement for an active immune system and the expression of an inflammatory response but this was not available to us. Immunodepression of concurrently infected mice could have been exploited to demonstrate that in the absence of an inflammatory response both species can co-exist together. However, the combination of immunodepression and two relatively pathogenic intestinal parasites is more than mice can tolerate at the doses that are required. Small-scale exploratory pilot experiments convinced us that this approach was not one which we wished to pursue further.
The reciprocal interaction, i.e. the effect of H. polygyrus on the expulsion of T. spiralis, was not an objective of this work but nevertheless our data confirm that there is an immunodepressive influence of H. polygyrus which was detectable in some experiments despite out attempts to minimize the reciprocal interaction. The majority of experiments were carried out with low doses of H. polygyrus, large enough for a reduction in the worm burden to be detected but low enough not to cause a major impairment of the response to T. spiralis. Behnke et al. (1978) demonstrated that expulsion of T. spiralis was most consistently impaired when the infection intensity of H. polygyrus exceeded 200 adult worms and in most of these earlier experiments doses of 200-400 worms were used. However, in spite of our precautions it was quite clear from some of the results we have presented here, as well as from the large amount of additional data arising from necessary control groups which we have not included, that the immunodepressive influence of H. polygyrus was still evident in concurrently infected