Host Selectivity and Distribution of Cassytha filiformis in the Coastal Bornean Heath Forests

Abstract

We investigated the host range of Cassytha filiformis L. in the heath forests using six 50-metre transects. Sixteen shrubs and tree species were infected by C. filiformis vines, including two exotic Acacia species. This paper also examined the density and vigour of C. filiformis when infecting the two most preferred and common hosts, the heath native Dillenia suffruticosa (Griff. ex Hook. f. and Thomson) Martelli, and the invasive Acacia mangium Willd. The results suggested that C. filiformis has higher vigour when infecting native hosts than in exotic A. mangium albeit being not statistically significant. The long thread-like stems of parasite were present at relatively high density when infecting A. mangium, regardless of the host conditions. We also assessed the functionality of the haustoria on both D. suffruticosa and A. mangium using histological methods. It was found that C. filiformis can establish a true haustorial endophytic connection with studied hosts. Under controlled conditions, C. filiformis pose as a possible candidate for a biological control agent of A. mangium to curtail the fast spreading of this introduced species in tropical Borneo.

Highlights.

• The investigation of the host selectivity of Cassytha filiformis in the heath forests using six 50-meter transects revealed that sixteen shrubs and tree species were infected by the parasitic vines, including two exotic Acacia species.

• C. filiformis exhibited higher vigour when infecting native hosts compared to exotic A. mangium and demonstrated relatively high density when infecting A. mangium, irrespective of host conditions.

• Using histological methods, C. filiformis can establish a true haustorial endophytic connection with A. mangium and D. suffruticosa.

INTRODUCTION

Throughout the course of evolutionary transitions, about 1% of angiosperms (Westwood et al. 2010) have adapted parasitism by acquiring resources from other plants via specialised organs of a morphological and physiological function called haustoria (Kuijt 1969; Yoshida et al. 2016; Teixeira-Costa & Davis 2021). Parasitic plants are often categorised by the extent of their host dependence (Heide-Jørgensen 2008). Facultative parasites are known to survive without a host for a certain period but would obtain their supply of water and/or nutrients when the opportunity arises. Alternatively, there are those that must require a host to live which are referred to as obligate parasitic plants. These plants are also recognised for their ability to photosynthesise (hemiparasites) or entirely non-chlorophyllous (holoparasites) (Musselman & Press 1995).

In terms of host preference, except for a few other specialists, most parasitic plants have a broad host range, especially when occurring in their natural habitat (Nickrent 2002). However, host specificity and the choice of hosts to infect ultimately depend on its accessibility and ability to locate hosts by selectively spreading towards or away from hosts, or by selectively penetrating host tissues upon contact through haustoria (Callaway & Pennings 1998; Runyon et al. 2006; Marquardt & Pennings 2010; Facelli et al. 2020).

Cassytha of the subfamily Cassythoideae is the only parasitic genus in the Lauraceae family (Awang et al. 2018). Cassytha filiformis Mill. is the sole pantropical species with wide global distribution in Asia, Africa, and tropical and subtropical America (Sastri 1962). It is a perennial hemiparasitic vine that infects its hosts by attaching to their stems. The generalist Cassytha has a relatively large and well-documented host range (Zhang et al. 2022). Despite the availability of hosts in the field, the obligate C. filiformis strands are often seen parasitising on only certain host species thus demonstrating the parasites’ preferential behaviour as highlighted by Koch et al. (2004) and Facelli et al. (2020). A common trait among generalists, the varying level of infection is also an indication of the mechanism of either active parasitism or a possible resistance on hosts (Kelly 1992) which could be examined by investigating the host stem histology and its anatomical response to the penetrating haustoria (Zhang et al. 2022). For instance, in a study by Facelli et al. (2020), Acacia myrtifolia was reported to exhibit resistance against the infection of Cassytha pubescens despite the presence of a firmly attached haustorium due to the lack of developed vascular connections. Under histological methods, the thickening cortical tissue of native species A. myrtifolia was observed thus preventing the parasite from forming true functional haustoria.

Cassytha are often seen along the coasts, sprawling on various host species at beaches around the world (Furuhashi et al. 2016). This is also a typical occurrence in Brunei where C. filiformis is abundant along the coasts (Rosli 2014; Tennakoon et al. 2016). Other than the preliminary list of hosts from an opportunistic field survey by Tennakoon et al. (2016), the study of host specificity in C. filiformis is lacking in Southeast Asia.

Despite accounting for 1% of the country’s forest cover, most of tropical Brunei’s coastlines are covered by a characteristic forest type known as heath forest. Bornean heath forests, locally referred to as Kerangas which means an area where rice cannot grow in the native Iban language (Davies & Salim 1999; Jambul et al. 2020), are mainly attributed to the highly acidic and low nutrient soils, and often inhabited by plant species with unique adaptive features (Newbery 1991; Wong et al. 2015; Hattori et al. 2019).

Tropical heath forests, especially those in Borneo, are susceptible to drastic environmental changes and anthropogenic activities (Din et al. 2015; Jambul et al. 2020). Similarly in Brunei, drastic changes in the ecosystem in the last 30 years have altered the soil properties causing this unique forest to be sensitive to degradation, fire, and habitat fragmentation (Zoletto & Cicuzza 2022). This is further exacerbated by the subsequent growth of the invasive and exotic Acacia species (Jaafar et al. 2016; Tuah et al. 2020) resulting in the secondary development of the now-threatened tropical heath (Kerangas) forest.

Much of the current host, C. filiformis studies looked into areas of its bioactivity (e.g., Abubacker et al. 2005; Armenia et al. 2015; Agbodjento et al. 2020; Umedum et al. 2020), physiology (e.g., Mukhtar et al. 2010; Mahadevan & Jayasuriya 2013; Balasubramanian et al. 2014; Furuhashi et al. 2021) and phylogeny (Wu et al. 2017; Zhang et al. 2020), while there are only few that discussed the effect of the stem hemiparasite on different hosts of a particular ecosystem (Kokubugata & Yokota 2012; Prider et al. 2009; Cai et al. 2020).

We present the first study on the host selectivity of C. filiformis in the threatened tropical Bornean heath forests. We examined:

1. Host range parasitised by C. filiformis using the transect method.

2. The impact of infection on hosts’ vigour relative to the density and vigour of the hemiparasite stem strands.

3. The anatomy of the haustorial interface of selected hosts to determine its functionality.

MATERIALS AND METHODS

Study Site

The study was conducted in the secondary heath (Kerangas) forests along the coastal highway (from 4°57’59.99°N, 114°52’33.531°E to 4°59’6.22°N, 114°54’1.472°E), within ca. 5 km off the coast of Brunei Darussalam from July to August 2021. Heath forests in Borneo are characterised by aseasonal lowland tropical rainforests that develop predominately on podzolised, highly acidic, sandy soils with relatively low macronutrients (Ghazoul & Sheil 2010; Jaafar et al. 2016; Ibrahim et al. 2020). Brunei has a tropical equatorial climate with average temperatures of 25.5°C and 28.9°C during the night and day throughout the year and total rainfall of 3815.1 mm in 2021 (Brunei Meteorological Service, unpublished data).

In the study area, the secondary heath forests are inhabited by a co-occurring composition of native species, such as Buchanania arborescens, Callophyllum inophyllum, Dillenia suffruticosa, Elaeocarpus mastersii, Melastoma malabathricum and Ploiarium alternifolium, and the invasive species of Acacia mangium, A. auriculiformis and A. holosericea (Tuah et al. 2020). C. filiformis are also observed infecting certain host plants. These species exist within the vicinities of settlements and urban developments (Fig. 1; see also Jambul et al. 2020).

Figure 1.

Map of Borneo Island (left) and the locations of the six transect surveys (T1 to T6) within the coastal heath forests of Brunei Darussalam (right).

Field Transect Survey

We established six 3 m × 50 m belt transects in July 2021 with ca. 0.5 m–1.0 m from the edge of the tropical heath forests. Within each transect, every individual of woody dicot species (i.e., shrubs and trees) with a height ca. 0.5 m and taller was recorded as “frequency of observation”, based on the methods employed by Kokubugata and Yokota (2012). The voucher specimens of the observed plants within the transect areas were collected for identification and confirmation at the Brunei National Herbarium (BRUN). Voucher specimens were deposited in the Universiti Brunei Darussalam Herbarium. To study the impact of infections on the two host plants with the highest frequencies of observations within these transects were selected, which are Acacia mangium Willd and Dillenia suffruticosa (Griff. ex Hook. f. and Thomson) Martelli.

Acacia mangium Willd. (hereafter Acacia) is a fast-growing leguminous tree species native to Australia and was introduced to Brunei in the late 1980s to mitigate soil erosion and as a timber plantation tree species (Osunkoya & Damit 2005; Ismail & Metali 2014; Jambul et al. 2020). It was then learnt that Acacia trees thrive in disturbed heath forests, especially since their seed dormancy is well-adapted to the recurring fire episodes and possesses the ability to fix nitrogen directly from the atmosphere (Jambul et al. 2020; Tuah et al. 2020). Osunkoya and Damit (2005) reported that Acacia could easily outcompete native plants such as Melastoma beccarianum under disturbed and degraded conditions, which eventually transform these habitats into nearly monospecific stands. Dillenia suffruticosa (Griff. ex Hook. f. and Thomson) Martelli. (hereafter Dillenia) is an important native pioneer shrub that may significantly impact the secondary succession of tropical forests (Rosli 2014). It is commonly distributed in disturbed areas, especially along roadsides and forest edges. Laboratory investigations have shown that Dillenia has anti-fungal, anti-bacterial and anti-cancer properties (Muliawan 2008; Armania et al. 2013; Goh et al. 2017).

The visual assessment of the host plants’ vigour and Cassytha cover were classified according to Prider et al. (2009). The vigour of Cassytha on each shrub was scored as “high” (actively growing, green stems), “low” (stems are partly dead and no active growth visible) or “dead” (no green stems). In our investigation, Cassytha cover was qualitatively scored as low, medium, high, and very high density. Low density infections covered <25% of the host where Cassytha was usually present as a few stems only, and medium density infections covered <50% of the host plant. High density infections covered <75% of the host, with Cassytha growing in entwined auto-parasitising strands to dense coiling mats. Very high density of Cassytha entailed the host plant being almost completely shrouded by the parasite, which can seem to deprive the hosts of sunlight.

Hosts’ growth condition or vigour was qualitatively scored as good, fair, poor and dead. “Good” hosts are when more than 90% of the individual plant is still alive where all or most of the leaves are green and intact. “Fair” host plants are 50% to 90% alive where some stems or leaves of hosts are dead or discoloured. Host plants that are mostly (<50%) dead or discoloured are scored as “poor”. Hosts are considered “dead” when all leaves are dead or discoloured. Cassytha infection was scored as present only when haustoria were observed on the plants within the transect areas. Chi-squared tests for independence were used to determine if there was a significant association between: (1) Cassytha vigour (i.e., High, Low, Dead) and its hosts; (2) Cassytha density (i.e., Very High, High, Medium, Low) and its hosts; (3) Cassytha density and the conditions of Acacia (i.e. Good, Fair, Poor, Dead); and (4) Cassytha density and the conditions of Dillenia (i.e., Good, Fair, Poor, Dead). Statistical analysis was conducted using R statistical programme version 4.1.3 RStudio (R Core Team 2022).

Haustorial Anatomy

Developing and attached mature haustoria on the selected hosts, Acacia and Dillenia, were fixed in an ethanol and xylene series as described in Tennakoon and Cameron (2006) and embedded in wax blocks with the haustorial interface arranged longitudinally. Using a microtome (Shandon Finesse ME+ Thermo Electron Corporation, Cheshire, UK), 10 μm–20 μm thick sections were prepared and placed onto glass slides. The thickness of the sections was based on the hardness of the host stems. Young and soft host stems were preferable in this experiment to ease the microtome process. Waxed sections were de-waxed and rehydrated prior to staining with 1% Toluidine Blue. Histological sections were examined under a light microscope (Leica DM2500 Microsystems CMS GmbH, Wetzlar, Germany). Images were acquired using a digital camera (Olympus DP73, Tokyo, Japan) using CellSens imaging software (Version 1.9, Olympus, Tokyo, Japan).

RESULTS

Host-Parasite Associations

A total of 336 individual dicotyledonous plants (see Appendix) were sampled from the six transect areas, where 99 individuals (29.5%) were found infected (Table 1). A total of 17 species from 16 genera and 15 families were recorded as host species. Buchanania arborescens, Dillenia suffruticosa, Elaeocarpus aff. mastersii, Nepenthes gracilis, Pouteria obovata, Psychotria sarmentosa, Rhodomyrtus tormentosa and Timonius flavescens were the native heath or Kerangas species identified (Coode et al. 1996; Tuah et al. 2020). Two invasive, introduced plant species, Acacia mangium and Acacia auriculiformis, were common and frequently observed within the study sites. Other than these two, the host species in Table 1 are native to Brunei (Coode et al. 1996; Zamri & Slik 2018; Tuah et al. 2020) and they are common to secondary forests of Brunei (Coode et al. 1996).

Table 1.

Summary of host plants from the six 3 m × 50 m transect surveys. The family and species names are arranged according to the frequencies of observations.

Host plants		Frequency of observationa
Family	Species
Dilleniaceae	Dillenia suffruticosa (Griff. ex Hook.f. and Thomson) Martellib	25
Fabaceae	Acacia mangium Willdc	19
Fabaceae	Acacia auriculiformis A. Cunn. ex Benthc	16
Melastomataceae	Melastoma malabathricum L.	10
Euphorbiaceae	Endospermum diadenum (Miq.) Airy Shaw	5
Nepenthaceae	Nepenthes gracilis Korth.b	5
Lamiaceae	Vitex pinnata L.	4
Elaeocarpaceae	Elaeocarpus aff. mastersii Kingb	3
Anacardiaceae	Buchanania arborescens (Blume) Blumeb	2
Malvaceae	Commersonia batramia (L.) Merr.	2
Rubiaceae	Timonius flavescens (Jacq.) Bakerb	2
Casuarinaceae	Casuarina equisetifolia L.	1
Euphorbiaceae	Macaranga tanarius (L.) Müll.Arg.	1
Myrtaceae	Rhodomyrtus tomentosa (Aiton) Hassk.b	1
Myrtaceae	Syzygium acuminatissimum (Blume) DC.b	1
Rubiaceae	Psychotria sarmentosa Blumeb	1
Sapotaceae	Pouteria obovata (R.Br.) Baehnib	1

Notes:

^abased on Kokubugata and Yokota (2012);

^bnative Kerangas species;

^cintroduced species.

Dillenia and Acacia were the two host plants with the highest frequency of observations (Table 1) and were selected to assess the in-situ effect of the Cassytha infection. The vigour of Cassytha while infecting the selected hosts was assessed in Fig. 2. Cassytha stems had higher vigour, i.e., better health while infecting the native Dillenia than that in Acacia, with more than 80% growing healthily in the former. It was found that there was higher mortality in Cassytha when infecting Acacia (16.7%) than that with Dillenia (4.0%). Chi-squared test was used to determine if there was a significant association between Cassytha vigour and the hosts. There was not a statistically significant association between the two variables (χ²(2, N = 43) = 2.24, p = 0.32).

Figure 2.

Impact of increasing Cassytha filiformis vigour on the two host species Acacia mangium and Dillenia suffruticosa. The vigour of Cassytha on each shrub was scored as “high” (actively growing, green stems), “low” (stems are partly dead and no active growth visible) or “dead” (no green stems).

The percentage of both host plants infected by various Cassytha densities is represented in Fig. 3. Despite the healthy growth of Cassytha on Dillenia, there was a higher infection density in the introduced species, with 27.8% and 22.2% of Acacia infected by high density and medium density of Cassytha, respectively. About 72% of Dillenia were infected by low density of Cassytha. None of the Dillenia and Acacia were infected by the very high density of Cassytha. Chi-squared test was also used to determine if there was a significant association between the increasing Cassytha density and the hosts. There was not a statistically significant association between the two variables (χ² (2, N = 43) = 5.06, p = 0.08).

Figure 3.

Impact of increasing C. filiformis densities on the two host species A. mangium and D. suffruticosa. Cassytha cover was qualitatively scored as low, medium, high and very high density. Low density infections covered <25% of the host where Cassytha was usually present as a few stems only, and medium density infections covered <50% of the host plant. High density infections covered <75% of the host, with Cassytha growing in entwined auto-parasitising strands to dense coiling mats. Very high density of Cassytha entailed the host plant being almost completely covered by the parasite.

Fig. 4 illustrates the health conditions or vigour of Acacia mangium and Dillenia suffruticosa with respect to the density of C. filiformis infection. In general, the virulence of Cassytha was high when host plants were healthy. However, the hemiparasite did not parasitise on Dillenia of lower vigour. Their preference was rather indifferent when infecting the introduced species where “poor” Acacia plants were parasitised by Cassytha. Chi-squared test was run to determine if there was a significant association between the increasing Cassytha densities and the growth conditions of hosts. There were no statistically significant associations between the two variables for both Acacia and Dillenia i.e., (χ² (6, N = 94) = 11.69, p = 0.07) and (χ² (3, N = 75) = 4.78, p = 0.19), respectively.

Figure 4.

Frequency histograms of the proportions of: (A) A. mangium, and (B) D. suffruticosa, in different growth conditions when infected by C. filiformis of increasing density levels. Hosts’ growth condition or vigour was qualitatively scored as good, fair, poor and dead. “Good” hosts are when more than 90% of the individual plant is still alive where all or most of the leaves are green and intact. “Fair” host plants are 50% to 90% alive where some stem or leaves of hosts are dead or discoloured. Host plants that are mostly (<50%) dead or discoloured are scored as “poor”. Hosts are considered “dead” when all leaves are dead or discoloured.

Histology of Haustoria Formation

Sections that were complete (intact) and best represent the behaviour of the haustoria are presented in Figs. 5 and 6. The haustorial endophytes had successfully reached and penetrated the vasculature of the host stems of D. suffruticosa (Fig. 5). The haustoria of C. filiformis appeared to grow into the host tissue in a wedge-like shape endophyte (E), mostly in direct contact with the host xylem (HX) (Fig. 5a). The presence of vascular core (V) was visible in middle section of the endophyte with the relatively high observation of xylem tissues. Another section of the same host – parasite (PX) association has shown direct luminal contact (Fig. 5b, PX, HX) with host xylem within the vascular core of the haustoria. Few cells of the endophytes were darkly stained, thus creating dense tissue (DT) in Fig. 5a. High nucleic-structures of hyaline body (HB) were present in the endophyte.

Figure 5.

Detailed anatomy of the haustorial interface of Cassytha filiformis with Dillenia suffruticosa at (a) ×4 magnification, and (b) at ×40 magnification, highlighting direct lumen-to-lumen xylem connections between the xylems of the host (HX) and parasite (PX). H = haustoria; P = host stem pith; PS = parasite stem; E = endophyte; HX = host xylem; PX = parasite xylem; I = interface between host and parasite; V = vascular core; DT = darkly stained tissue; CL = collapsed layer; HB = hyaline body.

Figure 6.

Detailed anatomy of the haustorial interface of Cassytha filiformis with Acacia mangium at (a) ×10 magnification (b) ×20 magnification, particularly a section of the haustorial endophyte. H = haustoria; P = host stem pith; PS = parasite stem; DX = differentiated xylem; E = endophyte; HX = host xylem; I = interface between host and parasite; IV = initial vascular core formation; CL = collapsed layer; HB = hyaline body.

As with the haustoria of C. filiformis on A. mangium in Fig. 6, the endophyte seemed to have spread around the host vascular structure creating an ellipsoidal flattened disc increasing the surface area of contact (Fig. 6a, I). The mass differentiating parenchyma cells running through the middle section of the endophyte indicate the initial development of the vascular core (Fig. 6a, IV). While the initiation of vascular core is yet to be present in Fig. 6a, differentiated xylem (DX) within the endophyte is evident in a different histological section (Fig. 6b). The presence of HB is also visible. The wedge-like endophytic growth of the parasite within the host tissue is also observed in other haustorial sections. This may be due to the relative thickness of D. suffruticosa stems, i.e., ca. 1.5 cm in comparison to the stems of A. mangium (ca. 0.5 cm–1.0 cm).

DISCUSSION

This study has shown the wide host specificity range of the hemiparasitic C. filiformis, thus exhibiting its generalist nature. This is evident in their unselective behaviour in infecting various host species, including the previously unrecorded grasses and fern species. The two most common host species for C. filiformis were A. mangium and D. suffruticosa. Although the parasite showed a slight preference for Dillenia, Cassytha thrives to high densities on both Acacia and Dillenia.

The results also demonstrated that under very high Cassytha density, a good Acacia stand exists. This may be because of the in-situ nature of the study where the age of the infection was not considered. The healthy Acacia was perhaps just newly infected, and the negative physiological effect of the infection was not apparent yet. Since every individual plant of height ca. 0.5 m and taller was recorded for this investigation, the age of the host plants is also highly variable. This potentially affects how the hosts respond to the parasitic infection. Nonetheless, host susceptibility to infection and the virulence of the parasite were greater in the introduced host than in the native host. This is a similar pattern observed in the parasitism of Cassytha pubescens on Leptospermum myrsinoides and Cytisus scoparius, a native and introduced species to Australia (Prider et al. 2009).

The soils of the threatened Kerangas forest are high in nitrogen and have always been negatively affected by Acacia trees which are invasive nitrogen-fixing legumes in Brunei (Tuah et al. 2020). A study by Yusoff et al. (2019) reported that there was a significantly higher concentration of total N in leaf litters in an Acacia-invaded Kerangas forest, suggesting that the invasive Acacia has further decreased the naturally poor soil nutrients. Non-fixers parasitic plants are likely to infect nitrogen-fixing hosts (Press et al. 1993; Seel & Press 1993) because hosts with high nutrient content such as legumes are often preferred (Matthies 1996; Pate & Bell 2000; Pennings & Callaway 2002) thus making Acacia spp. the highly favoured candidates as hosts. Recent nutrient studies on Cassytha hosts by Rosli et al. (unpublished data) have shown that D. suffruticosa has a similar amount of total N content (14.12 mg/g) to that of A. mangium (14.71 mg/g). The total N content of D. suffruticosa was also found to be higher than in the native pioneer, Melastoma malabathricum (12.59 mg/g).

The preference for hosts with relatively high nitrogen content is attributed to the lack of means to perform active uptake of such nutrients. Thus, hemiparasites like Cassytha opt to take organic nitrogen and other mineral nutrients that are diverted from the host xylem sap via the haustoria, to promote growth and increase their own biomass. Another reason for Cassytha’s acquisition of host-derived organic nitrogen elements is that they potentially lack the symbioses for nitrogenase enzyme production which is essential in biological nitrogen fixation. However, this assumption warrants further confirmation.

It is imperative to note that nitrogen-rich plants have reduced growth performance and are more vulnerable to parasitic infections which can further impair their stressed conditions (Kelly 1992; Gehring & Whitham 1992; Jeschke et al. 1994; Matthies 1996; Jeschke & Hilpert 1997; Pennings & Simpson 2008). Once infected, the host performance worsens as parasites thrive with the nutrients obtained from the hosts (Prider et al. 2009). It is evident in this study where C. filiformis also infected A. mangium at “poor” condition.

Bioactivity compounds attributed to the host-parasite dynamics also play a role in host specificity, specifically in the attachment process. The induction of chemical molecular signals, germination stimulants and haustoria-inducing factors are some examples of the products (Okubamichael et al. 2011; Yoshida et al. 2016). However, further investigations involving studies of bioactive compounds are required to confirm this potential cause.

Studies on host preference also reported that there are plant traits that appeared to be manipulated to demonstrate that they directly affect parasite preferences or performance (Kelly 1992; Pennings & Simpson 2008; Marquardt & Pennings 2010). This may account for Dillenia being one of the highest infected host plants in this study. This is also evident in the high percentage of Dillenia infection by a low density of Cassytha. C. filiformis are reported to prefer woody host plants with soft, thin barks and periderm and those with low and much-branched (Werth et al. 1979; Buriyo et al. 2015); both physical traits that are present in Dillenia as a pioneer, woody shrub that tends to grow in dense thickets. This indicates that Dillenia presents as a more accessible host to Cassytha by acquiring the required metabolites without investing much effort in heavy infestations.

Another possible explanation for the preferential behaviour may be attributed to the availability of more suitable resources which they acquire through the direct lumen-to-lumen linkages of the endophytes of A. mangium and D. suffruticosa: Cassytha associations. Through light microscopy investigations, this study was able to demonstrate such connections in the Dillenia-Cassytha association. This could not be captured in the Acacia-Cassytha sections, despite the proximity of the endophyte to the host vascular structure and the presence of the differentiating xylems. Thus, to further confirm this observation, we suggest utilising fluorescent trackers to the host root or sampling the xylem and phloem of the host and parasite and comparing the solute compositions (i.e., sugars, organic acids or amino acids) via isotope labelling (Tennakoon et al. 1997; Hibberd & Jeschke 2001; Jiang 2004; Tennakoon & Cameron 2006; Facelli et al. 2020).

Host tolerance to Cassytha infection may contribute to the reduced impacts of the parasites (Prider et al. 2009), however resistance was not observed in this study since no cases of pseudo-haustorial connections were encountered. It is also important to note that the field survey conducted in this in situ study did not determine if Cassytha was also connected to other surrounding hosts that could have been supporting its growth.

The outcomes of this study suggest that C. filiformis is indifferent to the hosts they parasitise, irrespective of whether hosts are native or exotic hosts. This confirms that generalist parasites are able to infect hosts which have not co-evolve to adopt a resistance or defence strategy (Koch et al. 2004; Cirocco et al. 2016). However, based on the results which highlight the higher density of C. filiformis on the invasive A. mangium, C. filiformis could be considered an important biological controlling agent under well-controlled conditions to reduce further spread of alien invasive A. mangium in tropical Southeast Asia. This concurs with the biotic resistance hypothesis where parasitic plants may be candidates for “a cost-effective environmentally sustainable component of invasion management scheme” (Těšitel et al. 2020). Generally, species that are used for biological control have high host specificity to ensure that only the targeted species is affected by the introduction of the species into a system (Myers & Bazely 2003). In the case of the secondary heath forest, most tree stands consist of the fast-growing Acacia species, and those infected are often in poor conditions based on field observations.

The next question should investigate this parasitism’s effects on the hosts’ physiology. The physiological impacts of parasites on invasive species have a greater effect on host health, biomass, and fecundity than on native hosts (Prider et al. 2009; Cirocco et al. 2016; 2018). Physiological studies such as photosynthetic activities and nutrient analysis on this host-parasitic association would be able to explain the extent of the impact of parasitism on these hosts within this unique site.

CONCLUSION

C. filiformis exhibited low host-specificity with its wide range of hosts, irrespective of their nativity to the tropical heath habitat. This is illustrated in the well-established haustorial structures in both A. mangium and D. suffruticosa. However, employing better histological techniques, such as scanning electron microscopy (SEM), may illustrate detailed anatomical evidence to prove successful haustorial connections. Previous studies conducted on C. filiformis in Brunei suggested that the hemiparasitic vine has the potential to act as a biocontrol agent against invasive species. The outcome of this investigation has shown that even with high Cassytha vigour, infected hosts can still thrive and did not specifically fulfil the previous statement. This could also be a possible inkling of a co-existing behaviour of Cassytha to certain hosts. This would entail an intricate look at resistance genes in the host genomes. However, in the existing state of the heath forests in Brunei where natives are threatened to be outcompeted by the monodominant A. mangium, C. filiformis is a good candidate for a potential biocontrol agent. This is feasible under controlled conditions by careful monitoring and ensuring that the spread of the hemiparasitic vines is limited to the invasive Acacia species only.

Furthermore, there are several determining factors and experimental modifications to this study that could be included to further test the impact of Cassytha infection on these hosts such as host biomass and the environmental conditions, for example, ex-situ and greenhouse experiments where the growth of the parasites and their hosts are monitored. Nonetheless, the findings also indicate that Cassytha can still be used to reduce the spread of exotic weeds and invasive plants.

APPENDIX

List of 336 dicotyledonous individual plants that were sampled during the field survey at the six 50-m transects, noted with the host plants’ vigour and Cassytha density.

Site	Individual	Infected (/)	Host Plant Vigour	Cassytha Density	Genus	Species
T1	1	X	Good	-	Acacia	mangium
T1	2	X	Good	-	Acacia	mangium
T1	3	X	Good	-	Acacia	mangium
T1	4	X	Good	-	Buchanania	arborescens
T1	5	X	Good	-	Acacia	mangium
T1	6	X	Good	-	Acacia	mangium
T1	7	X	Good	-	Acacia	mangium
T1	8	X	Good	-	Acacia	mangium
T1	9	X	Good	-	Acacia	auriculiformis
T1	10	X	Good	-	Acacia	auriculiformis
T1	11	X	Good	-	Acacia	mangium
T1	12	X	Good	-	Acacia	auriculiformis
T1	13	X	Good	-	Acacia	mangium
T1	14	/	Good	High	Acacia	auriculiformis
T1	15	/	Good	High	Buchanania	arborescens
T1	16	/	Good	Low	Syzygium	acuminatissimum
T1	17	X	Good	-	Syzygium	acuminatissimum
T1	18	X	Good	-	Callophylum	inophyllum
T1	19	X	Good	-	Acacia	mangium
T1	20	/	Good	High	Acacia	mangium
T1	21	/	Good	Low	Dilennia	suffruticosa
T1	22	/	Good	Low	Dilennia	suffruticosa
T1	23	X	Good	-	Acacia	mangium
T1	24	/	Good	High	Acacia	mangium
T1	25	/	Good	Low	Timonius	flavescens
T1	26	/	Good	Low	Acacia	auriculiformis
T1	27	X	Good	-	Syzygium	sp.
T1	28	/	Good	High	Acacia	mangium
T1	29	/	Good	High	Acacia	mangium
T1	30	/	Good	Low	Pouteria	obovata
T1	31	/	Good	Low	Endospermum	Diodenum
T1	32	X	Good	-	Endospermum	Diodenum
T1	33	/	Good	Medium	Endospermum	Diodenum
T1	34	/	Good	Low	Acacia	mangium
T1	35	X	Good	-	Acacia	mangium
T1	36	/	Good	High	Elaeocarpus	mastersii
T1	37	/	Good	Medium	Elaeocarpus	mastersii
T1	38	/	Good	Medium	Acacia	auriculiformis
T1	39	/	Good	High	Elaeocarpus	mastersii
T1	40	/	Good	High	Acacia	mangium
T1	41	/	Good	Low	Endospermum	diadenum
T1	42	/	Good	Low	Endospermum	diadenum
T1	43	X	Good	-	Elaeocarpus	mastersii
T1	44	X	Good	-	Endospermum	diadenum
T1	45	X	Good	-	Endospermum	diadenum
T1	46	X	Good	-	Acacia	auriculiformis
T1	47	X	Good	-	Endospermum	diadenum
T1	48	X	Good	-	Endospermum	diadenum
T1	49	X	Good	-	Endospermum	diadenum
T1	50	X	Good	-	Elaeocarpus	mastersii
T1	51	/	Good	-	Timonius	flavescens
T1	52	X	Good	-	Acacia	mangium
T1	53	X	Good	-	Pouteria	obovata
T1	54	X	Good	-	Pouteria	obovata
T1	55	X	Good	-	Endospermum	diadenum
T1	56	X	Good	-	Dillenia	suffruticosa
T1	57	X	Good	-	Pouteria	obovata
T1	58	X	Good	-	Licania	splendens
T1	59	X	Good	-	Elaocarpus	mastersii
T1	60	X	Good	-	Pouteria	obovata
T1	61	X	Good	-	Buchanania	arborescens
T1	62	X	Good	-	Acacia	mangium
T1	63	X	Good	-	Licania	splendens
T1	64	X	Good	-	Dilennia	suffruticosa
T2	1	X	Poor	-	Acacia	mangium
T2	2	X	Poor	-	Acacia	mangium
T2	3	X	Poor	-	Acacia	mangium
T2	4	X	Poor	-	Acacia	mangium
T2	5	X	Poor	-	Acacia	mangium
T2	6	X	Fair	-	Acacia	mangium
T2	7	X	Fair	-	Acacia	mangium
T2	8	X	Good	-	Syzygium	incarnatum
T2	9	X	Good	-	Acacia	mangium
T2	10	X	Good	-	Maranthes	corymbosa
T2	11	X	Poor	-	Acacia	mangium
T2	12	X	Poor	-	Acacia	mangium
T2	13	X	Fair	-	Acacia	mangium
T2	14	X	Fair	-	Acacia	mangium
T2	15	X	Poor	-	Acacia	mangium
T2	16	X	Poor	-	Acacia	mangium
T2	17	X	Poor	-	Acacia	mangium
T2	18	/	Poor	Low	Acacia	mangium
T2	19	X	Fair	-	Acacia	mangium
T2	20	X	Good	-	Picrophloeus	belukar
T2	21	X	Poor	-	Acacia	mangium
T2	22	X	Poor	-	Acacia	mangium
T2	23	/	Poor	Medium	Acacia	mangium
T2	24	/	Poor	Low	Acacia	mangium
T2	25	/	Fair	Medium	Acacia	mangium
T2	26	/	Fair	Medium	Acacia	mangium
T2	27	/	Fair	Medium	Acacia	mangium
T2	28	/	Fair	Low	Acacia	mangium
T2	29	X	Fair	-	Acacia	mangium
T2	30	X	Fair	-	Acacia	mangium
T2	31	/	Poor	Low	Acacia	mangium
T2	32	X	Poor	-	Acacia	mangium
T2	33	X	Poor	-	Acacia	mangium
T2	34	X	Fair	-	Acacia	mangium
T2	35	X	Fair	-	Acacia	mangium
T2	36	X	Fair	-	Acacia	mangium
T2	37	X	Fair	-	Acacia	mangium
T2	38	X	Fair	-	Acacia	mangium
T2	39	X	Fair	-	Acacia	mangium
T2	40	X	Fair	-	Acacia	mangium
T2	41	X	Poor	-	Acacia	mangium
T2	42	X	Poor	-	Acacia	mangium
T2	43	X	Good	-	Acacia	mangium
T2	44	X	Good	-	Acacia	mangium
T2	45	X	Fair	-	Acacia	mangium
T2	46	X	Good	-	Acacia	auriculiformis
T2	47	X	Good	-	Acacia	mangium
T2	48	X	Fair	-	Acacia	mangium
T2	49	X	Fair	-	Acacia	mangium
T2	50	X	Poor	-	Acacia	mangium
T2	51	X	Good	-	Elaeocarpus	mastersii
T2	52	/	Fair	Medium	Acacia	auriculiformis
T2	53	/	Good	Low	Melastoma	malabathricum
T2	54	X	Poor	-	Acacia	mangium
T2	55	X	Fair	-	Acacia	mangium
T2	56	X	Fair	-	Pternandra	coerulescens
T2	57	X	Good	-	Buchanania	arborescens
T2	58	X	Good	-	Acacia	mangium
T2	59	/	Fair	Low	Psychotria	sarmentosa
T2	60	X	Fair	-	Acacia	mangium
T2	61	X	Fair	-	Acacia	mangium
T2	62	X	Fair	-	Acacia	mangium
T2	63	X	Fair	-	Timonius	flavescens
T2	64	X	Fair	-	Psychotria	sarmentosa
T2	65	X	Fair	-	Acacia	mangium
T2	66	/	Fair	Low	Timonius	flavescens
T2	67	X	Fair	-	Acacia	mangium
T2	68	X	Fair	-	Melastoma	malabathricum
T2	69	X	Fair	-	Melastoma	malabathricum
T2	70	X	Good	-	Endospermum	diadenum
T2	71	/	Good	Low	Buchanania	arborescens
T2	72	X	Good	-	Acacia	mangium
T2	73	X	Good	-	Timonius	flavescens
T2	74	X	Good	-	Dillenia	suffruticosa
T2	75	X	Fair	-	Acacia	mangium
T2	76	X	Good	-	Timonius	flavescens
T2	77	X	Good	-	Timonius	flavescens
T2	78	/	Good	Low	Melastoma	malabathricum
T2	79	X	Fair	-	Melastoma	malabathricum
T2	80	X	Good	-	Timonius	flavescens
T2	81	X	Good	-	Melastoma	malabathricum
T3	1	X	Good	-	Dillenia	suffruticosa
T3	2	/	Fair	Low	Vitex	pinnata
T3	3	/	Good	Medium	Dillenia	suffruticosa
T3	4	/	Good	Very high	Melastoma	malabathricum
T3	5	/	Good	Medium	Melastoma	malabathricum
T3	6	/	Fair	Low	Comersonia	batramia
T3	7	X	Good	-	Melastoma	malabathricum
T3	8	X	Good	-	Melastoma	malabathricum
T3	9	X	Good	-	Melastoma	malabathricum
T3	10	X	Fair	-	Comersonia	batramia
T3	11	/	Good	Very high	Comersonia	batramia
T3	12	/	Good	Medium	Macaranga	tanarius
T3	13	X	Fair	-	Comersonia	batramia
T3	14	X	Good	-	Dillenia	suffruticosa
T3	15	/	Good	Low	Acacia	mangium
T3	16	X	Good	-	Melastoma	malabathricum
T3	17	X	Good	-	Dillenia	suffruticosa
T3	18	/	Good	Low	Melastoma	malabathricum
T3	19	X	Good	-	Melastoma	malabathricum
T3	20	X	Good	-	Melastoma	malabathricum
T3	21	X	Good	-	Melastoma	malabathricum
T3	22	X	Good	-	Melastoma	malabathricum
T3	23	X	Good	-	Melastoma	malabathricum
T3	24	X	Good	-	Melastoma	malabathricum
T3	25	X	Good	-	Vitex	pinnata
T3	26	X	Good	-	Acacia	auriculiformis
T3	27	X	Good	-	Acacia	auriculiformis
T3	28	X	Good	-	Melastoma	malabathricum
T3	29	X	Good	-	Melastoma	malabathricum
T3	30	X	Good	-	Melastoma	malabathricum
T3	31	X	Good	-	Melastoma	malabathricum
T3	32	X	Good	-	Melastoma	malabathricum
T3	33	X	Good	-	Dillenia	suffruticosa
T3	34	/	Good	Low	Melastoma	malabathricum
T3	35	X	Good	-	Dillenia	suffruticosa
T3	36	X	Good	-	Dillenia	suffruticosa
T3	37	X	Good	-	Dillenia	suffruticosa
T3	38	X	Good	-	Dillenia	suffruticosa
T3	39	X	Good	-	Dillenia	suffruticosa
T3	40	X	Good	-	Dillenia	suffruticosa
T3	41	X	Good	-	Dillenia	suffruticosa
T3	42	X	Good	-	Dillenia	suffruticosa
T3	43	X	Good	-	Dillenia	suffruticosa
T3	44	X	Good	-	Acacia	mangium
T3	45	X	Good	-	Vitex	pinnata
T3	46	/	Good	High	Nepenthes	gracilis
T3	47	/	Good	High	Nepenthes	gracilis
T3	48	/	Good	High	Nepenthes	gracilis
T3	49	/	Good	Medium	Nepenthes	gracilis
T3	50	X	Good	-	Nepenthes	gracilis
T3	51	/	Good	Low	Nepenthes	gracilis
T3	52	/		Medium	Rhodomyrtus	tomentosa
T4	1	/	Good	Low	Acacia	auriculiformis
T4	2	X	Good	-	Glochidion	littorale
T4	3	/	Fair	Medium	Acacia	auriculiformis
T4	4	X	Good	-	Melastoma	malabathricum
T4	5	X	Good	-	Vitex	pinnata
T4	6	X	Good	-	Dillenia	suffruticosa
T4	7	X	Good	-	Acacia	mangium
T4	8	/	Good	Low	Acacia	mangium
T4	9	X	Good	-	Vitex	pinnata
T4	10	X	Good	-	Vitex	pinnata
T4	11	/	Good	High	Vitex	pinnata
T4	12	/	Fair	High	Vitex	pinnata
T4	13	/	Good	Low	Vitex	pinnata
T4	14	X	Good	-	Acacia	mangium
T4	15	X	Good	-	Endospermum	diadenum
T4	16	X	Good	-	Dillenia	suffruticosa
T4	17	X	Good	-	Acacia	mangium
T4	18	X	Fair	-	Vitex	pinnata
T4	19	X	Good	-	Melastoma	malabathricum
T4	20	/	Good	Low	Melastoma	malabathricum
T4	21	X	Good	-	Acacia	mangium
T4	22	X	Good	-	Melastoma	malabathricum
T4	23	/	Fair	Low	Endospermum	diadenum
T4	24	X	Good	-	Acacia	mangium
T4	25	X	Good	-	Cocos	nucifera
T5	1	/	Good	High	Melastoma	malabathricum
T5	2	/	Good	Low	Acacia	auriculiformis
T5	3	X	Good	-	Acacia	auriculiformis
T5	4	X	Good	-	Acacia	auriculiformis
T5	5	X	Good	-	Acacia	auriculiformis
T5	6	X	Good	-	Alpinia	aquatica
T5	7	X	Good	-	Melastoma	malabathricum
T5	8	X	Good	-	Melastoma	malabathricum
T5	9	X	Good	-	Dilennia	suffruticosa
T5	10	X	Good	-	Dilennia	suffruticosa
T5	11	X	Good	-	Dilennia	suffruticosa
T5	12	X	Good	-	Melastoma	malabathricum
T5	13	X	Good	-	Dilennia	suffruticosa
T5	14	X	Good	-	Melastoma	malabathricum
T5	15	X	Good	-	Melastoma	malabathricum
T5	16	X	Good	-	Melastoma	malabathricum
T5	17	X	Good	-	Elaeocarpus	aff. mastersii
T5	18	X	Good	-	Melastoma	malabathricum
T5	19	X	Good	-	Melastoma	malabathricum
T5	20	X	Good	-	Dilennia	suffruticosa
T5	21	X	Good	-	Melastoma	malabathricum
T5	22	X	Good	-	Melastoma	malabathricum
T5	23	X	Good	-	Acacia	auriculiformis
T5	24	X	Good	-	Melastoma	malabathricum
T5	25	X	Good	-	Melastoma	malabathricum
T5	26	X	Good	-	Melastoma	malabathricum
T5	27	/	Good	Low	Dilennia	suffruticosa
T5	28	/	Good	Medium	Acacia	auriculiformis
T5	29	X	Good	-	Elaeocarpus	marginatus
T6	1	X	Fair	-	Acacia	auriculiformis
T6	2	X	Fair	-	Acacia	auriculiformis
T6	3	/	Good	Low	Casuarina	equisetifolia
T6	4	X	Good	-	Melastoma	malabathricum
T6	5	X	Good	-	Dilennia	suffruticosa
T6	6	X	Fair	-	Acacia	auriculiformis
T6	7	X	Poor	-	Acacia	mangium
T6	8	X	Fair	-	Acacia	mangium
T6	9	X	Fair	-	Acacia	auriculiformis
T6	10	X	Good	-	Dilennia	suffruticosa
T6	11	X	Good	-	Dilennia	suffruticosa
T6	12	X	Good	-	Dilennia	suffruticosa
T6	13	X	Good	-	Dilennia	suffruticosa
T6	14	X	Good	-	Dilennia	suffruticosa
T6	15	X	Fair	-	Acacia	mangium
T6	16	X	Good	-	Acacia	mangium
T6	17	X	Good	-	Dilennia	suffruticosa
T6	18	/	Fair	Low	Dilennia	suffruticosa
T6	19	/	Fair	High	Acacia	auriculiformis
T6	20	/	Fair	Low	Acacia	mangium
T6	21	/	Fair	Medium	Dilennia	suffruticosa
T6	22	/	Good	Medium	Melastoma	malabathricum
T6	23	/	Fair	Low	Acacia	mangium
T6	24	/	Fair	High	Acacia	auriculiformis
T6	25	/	Fair	High	Acacia	auriculiformis
T6	26	/	Good	Medium	Melastoma	malabathricum
T6	27	/	Good	Low	Dilennia	suffruticosa
T6	28	/	Fair	High	Acacia	auriculiformis
T6	29	/	Good	Low	Dilennia	suffruticosa
T6	30	/	Fair	V.High	Acacia	auriculiformis
T6	31	/	Good	V.High	Acacia	auriculiformis
T6	32	/	Fair	Medium	Acacia	auriculiformis
T6	33	/	Fair	Low	Acacia	auriculiformis
T6	34	X	Fair	-	Dilennia	suffruticosa
T6	35	X	Good	-	Melastoma	malabathricum
T6	36	/	Fair	Medium	Dilennia	suffruticosa
T6	37	/	Good	Low	Dilennia	suffruticosa
T6	38	/	Good	h	Acacia	auriculiformis
T6	39	/	Good	Low	Dilennia	suffruticosa
T6	40	/	Fair	Low	Dilennia	suffruticosa
T6	41	X	Fair	-	Dilennia	suffruticosa
T6	42	X	Fair	-	Dilennia	suffruticosa
T6	43	/	Good	Low	Dilennia	suffruticosa
T6	44	/	Good	Low	Dilennia	suffruticosa
T6	45	/	Good	Low	Dilennia	suffruticosa
T6	46	/	Good	Medium	Dilennia	suffruticosa
T6	47	X	Good	-	Dilennia	suffruticosa
T6	48	/	Good	Low	Dilennia	suffruticosa
T6	49	X	Fair	-	Dilennia	suffruticosa
T6	50	/	Good	Low	Dilennia	suffruticosa
T6	51	X	Good	-	Dilennia	suffruticosa
T6	52	/	Good	Low	Dilennia	suffruticosa
T6	53	X	Fair	-	Dilennia	suffruticosa
T6	54	X	Good	-	Dilennia	suffruticosa
T6	55	X	Fair	-	Dilennia	suffruticosa
T6	56	X	Fair	-	Dilennia	suffruticosa
T6	57	X	Good	-	Dilennia	suffruticosa
T6	58	X	Good	-	Dilennia	suffruticosa
T6	59	X	Fair	-	Dilennia	suffruticosa
T6	60	X	Good	-	Dilennia	suffruticosa
T6	61	X	Good	-	Dilennia	suffruticosa
T6	62	X	Good	-	Dilennia	suffruticosa
T6	63	X	Fair	-	Dilennia	suffruticosa
T6	64	/	Good	Medium	Dilennia	suffruticosa
T6	65	/	Good	Low	Dilennia	suffruticosa
T6	66	X	Good	-	Dilennia	suffruticosa
T6	67	X	Fair	-	Dilennia	suffruticosa
T6	68	X	Fair	-	Dilennia	suffruticosa
T6	69	X	Good	-	Dilennia	suffruticosa
T6	70	/	Good	Low	Dilennia	suffruticosa
T6	71	/	Good	High	Dilennia	suffruticosa
T6	72	X	Good	-	Dilennia	suffruticosa
T6	73	/	Good	Low	Dilennia	suffruticosa
T6	74	/	Fair	Medium	Acacia	mangium
T6	75	X	Good	-	Acacia	mangium
T6	76	X	Good	-	Acacia	mangium
T6	77	X	Good	-	Dilennia	suffruticosa
T6	78	X	Fair	-	Acacia	mangium
T6	79	/	Good	Medium	Dilennia	suffruticosa
T6	80	X	Good	-	Acacia	mangium
T6	81	X	Good	-	Casuarina	equisetifolia
T6	82	X	Good	-	Casuarina	equisetifolia
T6	83	/	Good	Medium	Casuarina	equisetifolia
T6	84	X	Good	-	Casuarina	equisetifolia
T6	85	X	Good	-	Casuarina	equisetifolia