Infective larvae of Haemonchus contortus found from the base to the top of the grass sward

The resistance of gastrointestinal nematodes (GIN) of sheep to anthelmintic treatment has motivated researchers to seek alternatives to reduce the use of these drugs in sheep farming and decontaminate pastureland based on knowledge about the survival dynamics of larvae. The aim of this work was to evaluate the migration of the infective larvae (L3) of Haemonchus contortus at different times of the day, strata, and sward heights, with and without shade after the deposition of contaminated sheep feces. The grass species used here was Cynodon dactylon cv. Tifton 85 in four treatments: low sward height shade; low sward height sunshine; high sward height shade; and high sward height sunshine. The number of L3 recovered from the pasture at different times of the day did not differ. The highest number of L3 recovered was in shade. The number of L3 at different times and strata occurred uniformly, confirming that L3 remain in the same place after migrating from dung at the hottest times of the day. Infective larvae of H. contortus were able to migrate across all the strata regardless of the time of day in the summer season in humid subtropical climate.


Introduction
Gastrointestinal nematodes (GIN) can reduce the profitability of sheep farms (Charlier et al., 2014) by directly impacting the productive performance of flocks, diminishing growth, and affecting the finishing period of animals (Pinheiro et al., 2000). The anthelmintic resistance of nematode parasites in sheep due to intensive drug use (Salgado & Santos, 2016) is an important factor to be considered in their management.
Approximately 95% of the parasite population is present in pasture and only 5% in animals (Bowman et al., 2003). An understanding of the vertical migration of parasite larvae and of the grazing behavior of herbivores could contribute to improve pasture management, aiming to reduce L3 intake and the use of anthelmintics to control parasites (Silva et al., 2008, Tontini et al., 2015. However, the number of H. contortus L3 in Brachiaria decumbens grass forage strata remained relatively constant throughout the day. This indicates there is no particular period of the day when grazing sheep are at higher risk of infection (Silva et al., 2008).
Several management practices have been tested in the attempt to reduce the damage caused by GIN in animals. However, as yet there is no single established recommendation that can be applied worldwide, because variations in climate, soil, and vegetation can affect the development and migration of larvae (van Dijk & Morgan, 2011). Even so, some practices are widely recommended by technicians, such as animals having access to pasture only after the dew has evaporated (Costa et al., 2011). This recommendation makes sense if L3 die or descend to the base of the plant as the due evaporates.
In fact, the availability of L3 in the pasture depends on the presence of free water, rain, or fog and condensed dew (van Dijk & Morgan, 2011). Free water moistens feces, allowing L3 to migrate from fecal matter to the pasture. Moreover, the formation of a continuous film of water on grass blades favors the migration of larvae to the top of the plant (Amaradasa et al., 2010). Santos et al. (2012) found H. contortus L3 at the top of Brachiaria decumbens even with low relative humidity (autumn). The authors stated that the moisture film produced by dew sufficed for L3 to reach the highest parts of the plant. Tontini et al. (2015), who evaluated the distribution of L3 during summer in the various strata of the tropical grass Panicum maximum cv. Aruana, found larvae distributed evenly across the strata. On the other hand, some studies have shown that L3 concentrate in the lower strata in some seasons (Vlassoff, 1982). Rocha et al. (2007) found a higher concentration of Trichostrongylus colubriformis in the upper strata of Panicum maximum cv. Aruana and in the lower strata of Brachiaria decumbens in autumn. However, in spring, the same authors found larvae in the upper portions of both forage species. These findings demonstrate that the location of L3 cannot be pinpointed, given that their migration can be facilitated or hindered by a number of factors.
Allowing animals access to well-managed pastures so they can graze the upper parts of plants of higher nutritional quality (Marchesan et al., 2013), combined with adequate nutrition planning, can reduce parasitism (Almeida et al., 2020). Nutrition plays an important role in the immune response to GIN (Coop & Kyriazakis, 2001). Animals that received high protein diets were able to better withstand parasitism (Bricarello et al., 2005;Louvandini et al., 2006). This fact, allied to knowledge about the grazing behavior of herbivores (which graze approximately 50% of the top of plants in the first mouthful) has been used to reduce L3 intake (Cangiano et al., 2002).
Based on the above findings, the purpose of this work was to evaluate the migration of L3 of Haemonchus contortus in Tifton 85 Bermuda grass (Cynodon spp.), in shade and sunshine, to different pasture heights at different times of the day, strata, and days after deposition of contaminated feces.

Study site and climate data
The experiment took place during February 2017 (28 days) in Ponta Grossa, Paraná, Brazil (25º 05' 49" S, 50º 03' 11" W; altitude 990 m). The region's climate is humid subtropical (Cfb), according to the Köppen classification, with moderately hot summers and relatively cold winters. Meteorological data (temperature, humidity, solar radiation, rainfall and dew) were provided by the meteorological station located one kilometer from the experimental site. The presence of dew was observed through a variable obtained from the weather station called leaf wetness duration (LWD), which corresponds to the time during which condensation water droplets remain on the leaf surface (Sentelhas et al., 2008). Figure 1 illustrates the climatic variations that occurred during the experimental period.

Experimental design
The experimental design involved Cynodon dactylon var. dactylon cv. Tifton 85 Bermuda grass (summer pasture). The area used for the experiment was not intended for grazing. The design consisted of sixteen 16.2 m 2 plots, each of which was subdivided into 18 subplots of 30 x 30 cm. The subplots were delimited with wooden stakes and string. The plots were separated by 0.5 m gaps to allow for circulation of researchers on the site.
The area of the experiment was divided into two parts: with and without the presence of trees (shade and sunshine). The forage heights were defined as high (20 cm) and low (10 cm). Thus, the following treatments were applied: low sward height -shade (LSH -shade), LSH -sunshine, high sward height -shade (HSH -shade) and HSH -sunshine.

4/10
Haemonchus contortus: across the grass Deposition of feces The experimental plot was contaminated using feces obtained from two male sheep naturally infected with GIN eggs. The use of rams (male sheep) was approved by the Ethics Committee on Animal Use at the State University of Ponta Grossa (Protocol 006/2016). The collection of fecal matter started on 24 Jan 2017 and the feces were stored in a refrigerator at 10°C. On the day of deposition, all the collected feces were weighed, homogenized, and six counts of eggs per gram of feces (EPG) were randomly conducted on the total volume of feces, involving a total of 288 samples, at 2.5 grams of feces, each containing an average of 60,000 eggs, following the procedure described by Silva et al. (2008). All the plots were contaminated (in the center of the plots) on 31 Jan 2017, using one fecal sample (2.5 grams) on each plot.

Recovery of infective larvae from forage and feces
Before collecting fecal matter and grass to evaluate larval migration, the height of the grass was determined based on the average height measured at five random points in each subplot, using a sward stick (Barthram, 1985).
To evaluate the vertical migration of L3 on the sward, the grass was cut manually and divided into three strata: upper stratum (A), consisting of 50% of the initial grass height, followed by an intermediate stratum (B) 25% below stratum A, and a final cut close to the ground to collect the lower grass (C). The samples of pasture were cut to simulate the grazing horizons of the animals. Ruminants use the strategy of preferentially grazing the first layers of forage, in order to maximize the quality of the diet and intake rate, since the upper layers user have more and better quality grass blades (Baumont et al., 2004).
Samples of fecal matter and grass were collected manually from the subplots and stored in labelled plastic bags until they were processed in the laboratory. The L3 were extracted using the technique described by Baerman and modified by Ueno & Gonçalves (1998). After 24 h in a Baermann apparatus, the supernatant was removed and the sediment transferred to graduated tubes. Subsequently, the L3 recovered from the samples were identified and quantified (Keith, 1953). The results are expressed as L3 per kilogram of dry matter (L3/kg DM).
After the L3 extraction procedure, the grass and feces samples were dried at 60ºC for 72 h to determine the dry matter (DM) content.

Statistical analysis
The study used block randomization with a 6 x 2 x 2 factorial design of treatments with six replicates at each time point (06:00 a.m., 12:00 p.m., and 06:00 p.m.), two treatments of luminance levels (shade vs. sunshine), and two treatments of grass height (high vs. low).
Larvae recovery data were log transformed (Log (x+1)) to account for non-homogeneous variance and were analyzed using the generalized linear model (GLM) of Minitab ® version 18.1 software (Minitab Inc., 2017). The data were analyzed per day of collection. Only the recovery of L3 in fecal matter was compared between the collection days, in order to determine how long the feces remained a viable reservoir for larvae. Collection time, stratum, and sward height were included in the model. The interaction between time and stratum was analyzed. However, to make them easier to understand, they are expressed as arithmetic means ± standard deviation. The means were compared by the Tukey test with a 5% level of significance.

Results
Only Haemonchus contortus counts were used, since this was the most abundant species of nematode. In this study, larvae of the genera Haemonchus (93.1%), Trichostrongylus (4.6%), and Oesophagostomum (2.3%) were recovered from Tifton grass.

L3 recovery and weight of fecal matter
The recovery of L3 from feces at the evaluated times did not differ (p>0.05). In general, the feces recovered in different times presented 7,217±24,644 L3/kg DM; 6,251±28,345 L3/kg DM, 8,927±30,897 L3/kg DM at 6:00 am, 12:00 noon and 6:00 pm, respectively. The same applies to the treatments, i.e., with no differences in the number of larvae recovered from feces between treatments (p>0.05) on the different days of collection (Table 1).
On D7, more L3 were recovered from fecal matter (29,960±50,427 L3/kg DM) than on any other day (p<0.05; Table 1). With regard to the weight of feces, on D7 the weight of LSH -shade was lower than that of the other treatments (p<0.05). In this treatment, fecal matter underwent 90.6% degradation. The maximum and minimum temperatures on the day of collection were 28.1ºC and 19.9ºC, respectively. The climatic conditions on D7 were 11 mm of rain, 87% RH, and 240.8 cal/m 2 of solar radiation.
On D14, both LSH and HSH -shade treatments had lower fecal weights than treatments exposed to sunshine (p<0.05). As expected, the treatments with the lowest fecal weights were also those with the highest percentage degradation (91.12% and 89.36%, LSH -shade and HSH -shade, respectively). On this day, the number of L3 in fecal matter were already quite low (28.80±215.5 L3/kg DM - Table 1).
On D21, the LSH -shade presented lower fecal weight (p<0.05) and 92.24% degradation compared to the high treatments (sunshine and shade), but did not differ from the LSH -sunshine treatment. The number of larvae recovered on D21 was 82.60±612.9 L3/kg DM. Lastly, on D28, the high sunshine treatment resulted in higher fecal weight (p<0.05) than the other treatments, and the lowest degradation rate, i.e., 83.2%. On D28, larvae were no longer recovered from feces. There was no difference in the number of larvae recovered from the pasture at the different times of day (p>0.05). In general, 46±278 L3/kg DM, 105±762 L3/kg DM and 82±371 L3/kg DM were recovered at 6:00 am, 12:00 noon and 6:00 pm, respectively. No significant interaction was found between time and stratum (p>0.05).
Evaluating the days separately, there was a significant difference in the recovery rate of L3/kg DM between treatments and strata on D7 (p<0.05). No L3 were recovered in LSH -sunshine on D7 ( Table 2). On that day, the mean number of L3 recovered (180 L3/kg DM) from stratum C was higher than the mean number recovered from other strata ( Table 2). On the other days of collection, the recovery of L3/kg DM did not differ between plant strata (p>0.05).
The recovery of L3/kg DM from the pasture also differed significantly between treatments (p<0.05) on D14. On that day, treatments LSH -shade (324±1037 L3/kg DM) and HSH -shade (153±352 L3/kg DM) showed higher larvae recovery rates than treatments in sunshine ( Table 2). On D21 and D28, the number of L3 recovered from pasture, in both periods, was minimal (p>0.05).

Discussion
GIN larvae successfully migrated through all the plant strata, regardless of the time of day. Seven days after the fecal deposition in the environment, a fraction of the L3 were recovered in fecal pellets. The wet weather that prevailed from the day of contamination until the first collection (D7) favored the development of eggs to larvae due to frequent rainfall, moderate temperatures and high humidity (Figure 1). H. contortus eggs developed rapidly to L3 (three to four days) within a temperature range of 20ºC to 25.8ºC, like the conditions reported by Hsu & Levine (1977). Temperature, humidity, and rainfall in the first seven days after the deposition of feces were important factors for the development of the free-living stages of H. contortus (O'Connor et al., 2007;Reynecke et al., 2011).
This experiment showed that on D7, larvae were still present in feces and in the lower strata, since seven days did not suffice for the larvae to leave the feces and migrate long distances from the ground.
The presence of dew in the pasture was observed only in the early hours of the morning. This indicates that even after evaporation/reduction of dew, the larvae remain in the same place after migrating from the feces. Larvae need free water only to abandon the dung, but not to climb up grass blades (van Dijk & Morgan, 2011). For larvae to migrate through the canopy of the plant, the moisture of dew is sufficient (Santos et al., 2012). In addition, larvae that reach the highest parts of the grass are able to remain there due to anhydrobiosis, a mechanism used by GIN larvae to enable them to survive the absence of water (Lettini & Sukhdeo, 2006).
Larvae exposed directly to solar radiation are subject to high mortality rates (van Dijk et al., 2009;Silva et al., 2008). This fact was observed in our study on D7 in LSH -sunshine and D14 in LSH and HSH -sunshine, when low numbers of L3 were recovered. The sward height in full sunshine, especially HSSH -sunshine, was found to provide a more favorable microclimate than LSSH -sunshine, as it probably reduced the intensity of solar radiation reaching the base of the grass sward, enabling some larvae to survive for longer periods (Moss & Bray, 2006).
Significant numbers of L3 were recovered on D14, possibly due to high RH (88.3%) and frequent rainfalls recorded from the day of deposition up to D14. The rainfall and RH that prevailed in the period caused an increase in the humidity of the feces, leading to higher degradation and favoring the migration of larvae to the grass sward (Almeida et al., 2020). Although we did not measure the climatic variables in each treatment (with and without the presence of trees -shade), the shade certainly contributed to retain more moisture in the environment than the

Sward heights
The sward structure was evaluated within each day and differed over the experimental period. On D7, all the treatments differed (p<0.05 - Table 3). On this day, the LSH -sunshine (15.50±3.3 cm) had the lowest height and the HSH -sunshine (25.94±2.5 cm) the highest height. On D14, HSH -sunshine (21.02±2.3 cm) differed from the other treatments (p<0.05). HSH -shade differed from LSH -sunshine, but not from LSH -shade. This, in turn, did not differ from LSH -sunshine. D21 followed the same trend as D7, with differences between all heights, LSHsunshine (14.05±2.7 cm) had the lowest and HSH -sunshine the highest height (30.75±3.1 cm). Lastly, on D28, the HSH -sunshine (28.52±2.9 cm) was higher than all the other treatments, and the treatments in shade (high and low) did not differ from each other (p>0.05), but differed from the LSH -sunshine, which presented the lowest height (15.46±2.4 cm).

8/10
Haemonchus contortus: across the grass treatments in sunshine (Almeida et al., 2018). Faria et al. (2016), who evaluated integrated crop-livestock-forestry systems, also reported higher recovery rates of GIN larvae than in systems without the presence of trees (sunshine).
Fourteen days after deposition of the feces in the environment, L3 were already present in the upper stratum, it indicates that the larvae contained in feces on D7 successfully migrated to the top of the plant after leaving the dung. The presence of plant cell extensions called trichomes in different forage species can influence larval migration (Niezen et al., 1998). Trichomes can hinder the movement of L3, or facilitate their migration by accumulating dew moisture (Niezen et al., 1998;Rocha et al., 2007). In our experiment, we did not measure the number of trichomes, but Tifton grass contains trichomes, albeit in very small quantities (Ahmad et al., 2016). The paucity of trichomes, allied to the shade produced in treatments with trees, facilitated the migration of larvae to the top of the plant, providing a microclimate with ideal conditions for larval development. Marley et al. (2006) reported that red clover negatively affected the migration of H. contortus larvae due to the presence of trichomes on its leaves and stems.
There is a positive correlation between rainfall and the emergence of larvae in pastures, since rain probably favors the development and survival of larvae in the field (Amaradasa et al., 2010). However, in this study, some L3 may have been removed from the pasture by a heavy rain that dumped 48.8 mm of water over a one-hour period, three days prior to collection on day 21. This intense rainfall may have carried some L3 away from the pasture (Grønvold & Høgh-Schmidt, 1989), which could explain the low L3 recovery on days 21 and 28 after fecal deposition. In addition, the moderate temperatures and high relative humidity recorded throughout the experimental period enabled larvae to develop and migrate to the sward in the first 14 days. However, this rapid development of larvae also resulted in their lower survival time in the environment (Rocha et al., 2014;Carneiro & Amarante, 2008).
Parasitological studies that evaluate the survival and migration dynamics of gastrointestinal nematodes in their developmental stages are important but increasingly meager. Such studies involve time-consuming but lowcost methods that allow for the choice of forages with strategic potential in the overall control of gastrointestinal nematodes reduced use of anthelmintics, thanks to the lower occurrence of L3, and the high nutritional quality and availability of forage during the months of greatest nutritional demand of animals susceptible to nematode infections.
In our experiment, the presence of shade and pastures with greater sward heights in full sunshine provided better climatic conditions for larval development and migration. In addition, H. contortus L3 were able to migrate across all the strata regardless of the time of day in the summer season. Therefore, there is no evidence that could help establish grazing schedules when animals are at lower risk of becoming infected by reducing larvae-host contact in regions of humid subtropical climate.