Immunological relationships during primary infection with Heligmosomoides polygyrus (Nematospiroides dubius): the capacity of adult worms to survive following transplantation to recipient mice

SUMMARY Chronic primary infections with Heligmosomoides polygyrus (Nematospiroides dubius) are still relatively poorly documented, particularly in relation to the role of host resistance in limiting worm survival. In the present work the duration of infection with H. polygyrus was studied in CFLP mice given doses of infective larvae ranging from 50 to 500 L3. The least heavily infected (50 L3) group ceased egg production earliest (week 36) whereas eggs were still detected in the faeces of mice given 500 larvae in week 42. At autopsy (week 42) mice given 50 larvae had virtually lost their entire worm burden with 5 out of 11 mice still harbouring a single worm each. However, all the mice in the group given 500 larvae were still infected, the highest worm burden being 93. The concentration of serum IgGl and specific antibody was highest in mice given 500 larvae, but sera taken from mice with declining worm burdens 19–38 weeks post-infection did not contain detectable host-protective antibody. During the course of infection in CFLP mice, H, polygyrus sustained irreversible changes in its capacity for subsequent survival. Thus, adult worms transferred to naive mice 2, 7, 14, 30 or 36 weeks post-infection did not live longer than worms of a comparable age in the respective donor group. In contrast, primary infection worms taken from jirds in which expulsion is usually completed by 6 weeks post-infection, re-established in mice and survived considerably longer than in the group of donor jirds. These results were discussed in relation to the possible interactions between parasite senility and immunomodulation, and host resistance in limiting primary infections with H. polygyrus in mice and jirds.


INTRODUCTION
The parasitic nematode Heligmosomoides polygyrus is capable of long-term survival in most strains of laboratory mice following a single administration of infective larvae (Ehrenford, 1954;Behnke & Robinson, 1985;Keymer & Hiorns, 1986). This longevity of H. polygyrus contrasts with the relatively brief existence of Nippostrongylus brasiliensis and Trichinella spiralis, both of which are expelled by rodents within 5 weeks of infection (Miller, 1984). H. polygyrus is believed to evade host-protective immune responses through immunomodulatory activity which impairs both afferent and efferent arms of the intestinal immune system (Behnke, 1987), enabling the parasites to live for 8-10 months, considerably longer than related species in rodent hosts. Although little is known about the nature of the factors involved and about their influence on the immune system at the cellular and molecular levels, it is predicted that the expression of host protective responses in infected mice must be preceeded by resistance to the immunomodulatory evasive mechanisms of the parasite (Behnke & Parish, 1979).
Most strains of mice acquire some capacity for resisting H. polygyrus following exposure to repeated or abbreviated infection protocols (Prowse, Mitchell, Ey & Jenkin, 1979;Behnke & Robinson, 1985) and the mechanisms involved have been studied in several laboratories in recent years (Dobson & Cayzer, 1982;Jacobson, Brooks & Cypess, 1982;Pritchard, Williams, Behnke & Lee, 1983;Pentilla, Ey, Lopez & Jenkin, 1985). However, immunological events during the chronic primary infection, particularly in the terminal stages, 6-10 months after initial exposure to larvae, have been relatively neglected. Nevertheless, this may be a particularly crucial phase of the interaction between host and parasite because there exists the possibility that the strategies used by H. polygyrus to evade host immunity during the initial stages of infection cease to be effective in old worms, thereby allowing the host to remove the parasites through the expression of previously suppressed mechanisms. Alternatively, the loss of 8 to 10-month-old worms may simply ensue from senility, i.e. endogenous changes within the parasites which are not influenced by the host. A third possibility is that the worms do not undergo changes within the period of infection contributing to their demise but the host, through prolonged exposure to immunomodulatory molecules, develops counter-effective factors (e.g. antibodies) which enable hostprotective responses to be fully expressed.
In this paper, we report experiments which were carried out to determine whether the loss of adult H. polygyrus in the terminal stages of a chronic primary infection is mediated through the immunological activity of the host and is accompanied by irreversible changes to the parasite's capacity for survival. Worms from jirds (Meriones unguiculatus) were studied for comparison because, in contrast to mice, M. unguiculatus expel adult H. polygyrus through an intestinal immune response within 6 weeks of infection (Jenkins, 1977;Hannah & Behnke, 1982).

Animals
Randomly bred CFLP and syngeneic NIH mice were used in this study. Both strains of mice and jirds were bred in the department and housed under conventional animal house conditions with access to food and water ad libitum.

Parasite
The methods used to maintain H. polygyrus, infect mice and recover worms at autopsy have all been described previously (Jenkins & Behnke, 1977). Faecal egg counts were carried out as reported by Behnke & Parish (1979). A detailed description of the methods used to transplant adult worms was given by Behnke, Hannah & Pritchard (1983).

Measurement of immunoglobulins and antibodies in sera from infected mice
Immunoglobulin levels were determined by single radial immunodiffusion (Mancini assay) as described by . Goat anti-mouse subclass sera were purchased from Serotec (Bicester, U.K.) and diluted in phosphate-buffered saline (PBS) before addition to immunodiffusion plates. Immunoglobulin levels were calculated by reference to calibration curves prepared with mouse standards.
Antibodies specific for adult worm homogenate were detected using an ELISA assay. Adult homogenate in PBS was diluted in carbonate-bicarbonate buffer, pH9, to a final concentration of 400/tg/ml and the assay conducted as described by Ruitenberg, Steerenberg, Brosi & Buys (1979) using a urease-conjugated second antibody (Sera Lab. Crawley).
Immune mesenteric lymph node cells (IMLNC) were prepared as described by Behnke & Parish (1981) and . Recipient mice were given the cells in 05 ml of medium, 4 days before infection. Challenge infection larvae were exposed to 5 krad. of Cobalt 60 gamma-radiation as reported by Behnke, Parish & Hagan (1980). Infective larvae prepared in this way are particularly sensitive to protective responses by the host (Behnke & Parish, 1981) and, in comparison to normal larvae, the reduction of worm burdens in recipient mice following passive or adoptive transfer of immunity is markedly enhanced. Primary infection sera were stored at -20 °C until required and were administered to recipient mice as follows: 0 -5 ml on day 0, l -0 ml on day 1 and 0 -5 ml on day 3.

Statistical analysis of results
Faecal egg counts are presented as the mean value of 4 counts on each 1 g sample of faeces (e.p.g.). Worm counts are given as the mean worm burden + S.E.M. When applicable the non-parametric Mann-Whitney U test was used to compare groups for significant differences (Sokal & Rohlf, 1969) and a value of P < 0-05 was considered to be significant. Differences in the concentration of immunoglobulins were analysed by a Student's «-test (Sokal & Rohlf, 1969).

RESULTS
The duration of infection in mice given varying numbers of L 3 larvae Four groups of 12 CFLP mice (3 groups of male and 1 group of female mice) were infected with different numbers of L 3 and faecal egg counts were recorded until week 42 (Exp. 1). The 3 groups of male mice were given 500 (group A), 250 (group B) and 50 (group C) larvae respectively, whilst female mice (group D) received 250 larvae. Groups of 8 male (group E) and 8 female (group F) age-matched controls, which were left uninfected were also included in this experiment. The results are shown in Fig. 1 and Table 1. The least heavily infected group (C) became negative first, in week 36, but was followed within 5 weeks by groups B and D. The only group still producing parasite eggs albeit in very low numbers on the day when the animals were killed for worm counts was the most heavily infected group (A). Very few worms remained until week 42. Male mice given 50 larvae (group C) had virtually lost their entire worm burden, with 5 out of 11 mice still harbouring a single worm each. In group B only 2 mice were totally without worms, whereas all the mice in group A were still infected and the highest worm burden was 93.

IgG 1 and parasite specific antibody in CFLP mice with a primary infection
All the animals in the preceding experiment (Exp. 1) were bled at autopsy and individual sera were analysed for total IgG 1 and for parasite-specific antibodies (Table  1). IgG 1 levels in all the infected animals were significantly increased in relation to control groups and both IgGl and specific antibody levels were highest in group A which received the most larvae.
A second experiment (Exp. 2) was carried out in which a group of 6 to 8-week-old female CFLP mice was infected with 200 L 3 and was bled at weekly intervals for 56 days  after infection. A further sample of serum was obtained on day 210. Serum immunoglobulin isotypes were measured and the results are presented in Fig. 2. With the exception of IgA on day 14, only IgG 1 levels showed a significant change. These rose from a resting level of 2-1 mg/ml to a maximum of 5-0 mg/ml on day 21 and thereafter progressively declined to a concentration of 3-1 mg/ml on day 210. The possible existence of host-protective antibodies in long-term primary infection sera was investigated by passive and adoptive transfer experiments. A group of 120 female CFLP mice was infected with 250 L 3 and 30 mice were bled during weeks 0, 40000 e.p.g. The worm burdens in weeks 8 and 15 were 2645+145 and 222*4±8-5 respectively. By week 20 faecal egg counts began to decline, culminating in a value of 200 e.p.g. during week 38. The worm burden in mice killed during weeks 19 and 38 were 2256+140 and 9-2 + 4*9. The pool of serum obtained during week 19 was therefore taken immediately prior to worm loss and that during week 38 when 96*5 % of the worm burden had been eliminated.
Each of the 4 pools of serum was injected into 2 groups of 6 female NIH mice, alone or together with 9 x 10 7 immune mesenteric lymph node cells (IMLNC). All the mice were challenged with 200 5krad. irradiated L 3 and were killed 18 days later for worm counts. The mean number of worms recovered from the control group was 181*3 + 4-6. The group receiving IMLNC alone had 57-4+190 worms, reflecting a 6 8 3 % reduction in worm burden. All the groups of mice receiving serum alone had worm burdens indistinguishable statistically from those of the control group and the mice which were given IMLNC together with one of the serum pools had worm burdens comparable to the group given IMLNC alone (data not shown). It is clear from these results that the administration of serum from chronically infected mice had no effect on the challenge infection and did not interact synergistically with concurrently transferred IMLNC, irrespective of the time after infection when serum was taken (see Behnke & Parish, 1981). It is therefore unlikely that the loss of primary infections from CFLP mice is accompanied by sufficient host-protective antibody in the serum to be detected in this assay. Host-protective antibody in mice which develop resistance during repeated infection with H. polygyrus is reliably detectable by this technique .

The survival of adult worms transferred from mice harbouring a primary infection, into naive recipients
In order to determine whether the loss of parasites in mice was attributable to irreversible damage sustained by the organisms, a series of experiments was carried out in which worms were transferred into naive recipients at various times in relation to infection.
In Exp. 4, a group of 12 female CFLP mice were infected with 300 larvae of H. polygyrus. Fourteen days later adult worms from 6 of these mice were transplanted into a group of 6 naive female CFLP mice. The faecal egg production from both groups was monitored at fortnightly intervals until week 30. Worms from half of the remaining mice in each of these 2 groups were then further transplanted into 2 groups of naive recipients, and faecal egg counts from all 4 groups were monitored until no further eggs were detected.
It is clear from Fig. 3 that transplantation of worms on day 14 did not markedly alter the survival pattern of worms. Furthermore transplantation of worms during week 30, when egg counts were falling rapidly did not improve worm survival, and consequently egg counts continued to decline in all 4 groups This experiment was repeated with some modifications. A group of 12 female CFLP mice were infected with 400 larvae (Exp. 5) and the faecal egg production was monitored until week 30. At this time worms were transplanted from 6 mice in this group into a group of 6 naive female recipients. Two weeks earlier, i.e. week 28, a second donor group of 12 female CFLP mice was infected with 400 larvae of//, polygyrus and during week 30, when these worms were only 2 weeks old, half of the donor group was killed and their parasites were transplanted into 6 recipient mice. The experiment therefore comprised 4 groups, 2 were the donor groups infected in week 0 and week 28 respectively, and 2 were recipient groups one of which received 30-week-old worms whilst the other was concurrently given 2-week-old parasites.
The faecal egg output from transplanted 30-week-old worms declined at the same rate as that of the worms in the remaining mice from the original donor group (Fig. 4). However, the faecal egg counts from the mice infected in week 28 and from those receiving 2-week-old worms during week 30 remained constant for up to 15 weeks after transplantation, but were reduced by week 57. It is clear from these results that both   A final experiment (Exp. 6) was carried out along the same lines but incorporating further modifications. A donor group (A) comprising 30 female CFLP mice was infected with 300 larvae during week 0 and worms from this group were transplanted into naive recipient mice in weeks 7 (group B), 14 (group E) and 36 (group H). Two weeks prior to each transplantation additional groups of mice were infected (groups C, F and I) so as to provide concurrently 2-week-old worms for transplantation. Thus the survival of 7, 14-and 36-week-old worms was compared to that of 2-week-old worms on each occasion. Faecal egg counts were carried out on all the donor and recipient groups until no more eggs could be detected and the experiment was terminated during week 43. The experimental design is given in Table 2 and the results are presented in Fig. 5.
Faecal egg counts remained relatively steady in the original donor group (A) until  week 26, when counts began to fall steadily, but eggs were detected in the faeces until week 42. The 7-week-old worms from group A re-established in recipient mice (group B) but egg production only lasted for a further 26 weeks. Mice infected during week 5 (group C), however, showed a shorter course of egg production than expected, and faecal egg counts from group C and from group D, which received 2-week-old worms, did not extend beyond those of the original donor group A. The transplantation of 14-week-old worms to naive recipients (group E) did not result in prolonged egg production in relation to group A. Concurrently transplanted 2week-old worms (group G), produced fewer eggs, suggesting that establishment may have been poor, but faecal egg counts remained steady for 15 weeks in contrast to group E, before falling prematurely during week 32.
When 36-week-old worms were transplanted to naive recipients, faecal eggs were detectable for a further 2 weeks (group H). Concurrently transplanted 2-week-old worms, however, sustained high egg production until the experiment was terminated during week 43.

The survival of adult worms transplanted from jirds harbouring a primary infection,
into naive recipient mice Twenty-three jirds were infected with 400 larvae, and in 5 animals the course of infection was followed by regular egg counts. The remaining 18 jirds were killed in groups of 3-4 animals on days 10, 20, 25 and 31 and on each occasion 10 male and 10 female worms were transplanted from the donor jirds to each of 5 male CFLP mouse recipients. Faecal egg production was subsequently monitored in each group until day 70. The results are shown in Fig. 6.
Parasite eggs were first detected in jirds 10 days after infection, rose to a peak on day 14 and declined thereafter, until no more eggs could be detected on day 34. Worms established in all the recipient groups of mice and egg production continued steadily in each group until day 70. The mean number of worms recovered from groups of mice receiving worms from jirds on days 10, 20, 25 and 31 was 7-2 + 2-9, 8-8± 2-4, 2-9 + 0-9 and 8-3 + 1-5 respectively. Despite the erratic establishment in recipient mice, some adult parasites survived for more than 4 weeks, longer than in the donor group of jirds. Five remaining jirds examined for worms on day 70 were found to have expelled all their parasites.

DISCUSSION
Although H. polygyrus is known to have the capacity to cause long-lasting infections in mice, the duration of infection is variable and influenced by the strain of mouse and the intensity of infection (Prowse et al. 1979;Dobson, Sitepu & Brindley, 1985). Keymer & Hiorns (1986) found that the rates of decline in parasite numbers in MFI mice were similar irrespective of the infection dose. Dobson et al. (1985) reported lowintensity infections to be eliminated more rapidly than heavy infections in Quackenbush mice, and our results concur with the latter study. Thus male CFLP mice, initially given 50 larvae, lost virtually their entire worm burden by week 42, whereas the more heavily infected animals still retained significant numbers of adult worms at this time (Table  1). This is an important result because it provides a possible explanation for the data from subsequent transplantation experiments. The original donor groups were invariably given a heavy infection which lasted for at least 35 weeks. In contrast, 2week-old worms transplanted concurrently with 7-week-old worms failed to outlive the latter or the worms in the donor groups (Fig. 5). I t is possible that relatively few worms established in the recipient animals and consequently worm loss was initiated earlier.
The shorter duration of low intensity infections with H. polygyrus is compatible with the hypothesis that worm survival is dependent on the immunomodulatory activity of adult parasites (Behnke, 1987). Support for this interpretation is found in the results of experiments analysing dose-dependent variation in the onset of the expulsion of Trichinella spiralis (Bell, McGregor, Woan & Adams, 1983;Bell, Adams & Ogden, 1984;Wassom, Dougherty, Krco & David, 1984;Wakelin, Donachie & Grencis, 1985). The delayed rejection of T. spiralis from heavily infected animals has been attributed to the slow response of a population of T lymphocytes, controlled by H-2 and non H-2 genes, which seem to be particularly sensitive to the suppressive influence of heavy parasite burdens (Wakelin et al. 1985). If worm loss is dependent on the sensitivity of a particular population of lymphocytes, crucial t o the expression of host protective immunity, to immunomodulatory factors of both parasite species, low-intensity infections may pose less of an obstacle than heavy infections and hence expulsion would be initiated earlier.
It is clear from the results illustrated in Figs 3-5 that in no case was worm survival, as monitored by faecal egg counts, extended by transplantation of parasites into naive hosts. Therefore, it would appear that the worms undergo endogenous changes with increasing duration of infection or their evasion strategies become less effective, facilitating the expression of a protective response by the host. Although H. polygyrus may live for up to 46 weeks post-infection ( Fig. 1) in most of our experiments survival was briefer. This variation in longevity and the fact that some inbred mouse strains (e.g. SJL) expel worms within 2 months of infection, makes it unlikely that senility is the only factor involved, and suggests that the murine host makes an active contribution to parasite loss during primary infections. Whether the host component is immunological in nature is uncertain. Our experiments did not reveal major changes in immunoglobulin or antibody levels coinciding with worm loss but we cannot exclude the possibility that intestinal IgA or cellular responses intensified during the final stages of infection, causing irreparable damage to the parasites prior to their loss from the gut.
As the course of infection progressed, H. polygyrus were observed to undergo changes which were evident at both the light and electron microscope levels (details to be reported elsewhere). The most obvious feature was a gradual darkening of adult worm haemoglobin from an initial bright red to an almost dark brown colour in the terminal stages of infection. Furthermore, electron microscopical analysis revealed depletion of the gonads and damage to the intestinal cells and to the body wall comparable to the changes encountered in N. brasiliensis, T. spiralis, Strongyloides ratti and Haemonchus placei during the loss phase of these species from their respective hosts (Kennedy & Bruce, 1981;Moqbel & McLaren, 1980;Ogilvie & Hockley, 1968;Harness, Smith & Bland. 1973). T. spiralis and S. ratti have been shown to recover from damage sustained during the immune response, upon transplantation to naive hosts (Kennedy & Bruce, 1981;Moqbel & McLaren, 1980), but in N. brasiliensis damage appears to be irreversible (Ogilvie & Hockley, 1968) and the survival characteristics of//, polygyrus in the present work suggests that the same may be true of this species. The exact cause of the degenerative changes encountered in worms during expulsion has not been conclusively established, although antibody has been implicated (Jones & Ogilvie, 1971) and it has been proposed that worms driven into the suboptimal environment of the inflamed intestinal lumen may sustain damage through an inability to respire and feed normally under these circumstances (Lee, 1969;Love, Ogilvie & McLaren, 1975).
It is interesting that the loss of H. polygyrus from jirds was not accompanied by irreversible alterations in the capacity of parasites to survive after transplantation. Thus worms transferred from jirds into mice, even as late as day 31, re-established and continued egg production for at least 5 weeks, long after they would have been eliminated from the donor host. Infections in jirds are also subject to dose-dependent suppression, heavy worm burdens persisting longer than low intensity infections (Hannah & Behnke, 1982). However, the jird's immune system is not as susceptible to manipulation by the parasite as that of the mouse, since parasite survival is usually curtailed within 2 months of infection. Jirds are not a natural host for H. polygyrus in the wild (Hannah, 1983) and may not provide an environment which is entirely optimal in all respects; the growth of adult worms is slower initially than in mice although a comparable size is eventually attained (Behnke & Hannah, 1984;Jenkins, 1977). Under these conditions it is conceivable that expulsion can be effected without the prerequirement of severe damage. In contrast, in the mouse intestine, where presumably the environment is more favourable, the worms immunomodulate successfully and probably compensate for and repair initial damage. Mice therefore may be unable to remove the parasites until changes within the organisms are so severe as to markedly enhance their susceptibility to other effector mechanisms. The crucial question, however, is whether these changes are entirely endogenous or whether they are brought about by a first-step process, mediated through antibodies as has been suggested for N. brasiliensis (Jones & Ogilvie, 1971). It will now be interesting to determine whether H. polygyrus from SJL mice just prior to rejection, exhibit irreversible changes which increase susceptibility to a second-step expulsive mechanism and to establish whether the capacity of these mice to terminate adult worm infections so rapidly in relation to other strains is linked to a genetically determined insusceptibility to parasite-mediated immunodulation.