University of Hawai'i Press
  • Diet Analysis of Hawai‘i Island’s Blackburnia hawaiiensis (Coleoptera: Carabidae) using Stable Isotopes and High-Throughput Sequencing

Determining the diet of arthropods can be difficult due to their small size and complex food webs, especially in Hawai‘i, where knowledge of arthropod predator–prey interactions is sparse. The diet of the Hawai‘i Island-endemic carabid beetle, Blackburnia hawaiiensis Sharp (Coleoptera: Carabidae) is of particular interest because of its peculiar arboreal behavior and metathoracic flight wings. Our study objective was to determine the diet of B. hawaiiensis in replicated, geographically separated locations by using two different yet complementary laboratory techniques: natural abundance stable isotope analysis (SIA) and high-throughput sequencing (HTS). Overall, B. hawaiiensis had a greater average δ15N and similar δ13C compared to the other arthropods sampled in this study and HTS data revealed Diptera and Lepidoptera sequences in the beetle’s gut contents. These results are consistent with B. hawaiiensis being classified as a generalist predator. The combination of SIA and HTS are important methods for determining the diet of species within complex food webs, particularly for species that are difficult to observe in nature.


food web, metabarcoding, Hawai‘i, carabid, B. hawaiiensis

Understanding the diet of organisms can elucidate conservation needs and inform management. However, trophic relationships are often difficult to observe, especially when considering small, rare, or cryptic species (Gomez-Polo et al. 2015). Traditional diet studies often include direct observation of scat or gut contents, although these techniques are limited to organisms that consume indigestible structures leaving solid, identifiable remains (Hoogendoorn and Heimpel 2001). Specifically, arthropod diets and trophic positions are difficult to assign not only because of their small size but also because many are generalists, therefore creating complex food webs (Polis et al. 1989). Additionally, generalist predators can partake in cannibalism and intraguild predation, eating not only herbivores but also other predators competing for the same resources (Sabelis 1992, McNabb et al. 2001, Mestre et al. 2013).

With complex food webs, stable isotope analysis (SIA) can be a helpful tool for characterizing trophic relationships among organisms (Mestre et al. 2013). Particularly, δ15N and δ13C are well documented to be useful for assessing trophic position, where the heavier isotope accumulates with [End Page 245] successive trophic levels (Cabana and Rasmussen 1996, Post 2002, Johnston et al. 2018, Kennedy et al. 2018, Kennedy et al. 2019). In order to properly estimate trophic positions in a given food web the establishment of an isotopic baseline in that given system is necessary but often difficult to discern (Post 2002, Kristensen et al. 2016). In addition, the variability in measurements in species-rich, complex food webs can make it difficult to assign trophic levels (Mestre et al. 2013). Therefore, it can be advantageous to employ a complementary technique such as the use of genetic metabarcoding of gut contents, which can yield higher prey resolution.

The ability to elucidate food webs through genetics has greatly improved with recent technological advancements (Mestre et al. 2013, Piñol et al. 2014b, Gomez-Polo et al. 2015). Metabarcoding, the genetic study of environmental samples utilizing universal primers and high-throughput sequencing (HTS), facilitates the identification of complex mixtures of DNA sequences (Shokralla et al. 2012). HTS has become a powerful tool for ecologists, allowing the examination of dietary breadth without the use of species-specific primers (Gomez-Polo et al. 2015). For example, Piñol et al. (2014a) used HTS to determine the diet of linyphiid spiders, Gomez-Polo et al. (2015) determined that of a hemipteran, and Krehenwinkel et al. (2017a) determined the diet of grass spiders. Although effective for determining potential prey items, HTS has limitations such as sequencing error and incomplete reference libraries (Deagle et al. 2013, Piñol et al. 2014b, Thomas et al. 2014). Because of the limitations of HTS, combining SIA for a time-integrated trophic structure may be advantageous, particularly for complex arthropod food webs (Vander Zanden et al. 1997).

The integrated approach of using both SIA and HTS allows for a better understanding of the feeding ecology of organisms that are difficult to rear in the lab and observe in nature. Employing these two techniques provides complementary information with two lines of diet evidence at different resolutions (Carreon-Martinez and Heath 2010, Whitaker et al. 2019). Stable isotope measurements are the result of a natural accumulation over time, while HTS identifies prey eaten within a few days or hours prior to collection.

Blackburnia hawaiiensis Sharp (Coleoptera: Carabidae) is a Hawai‘i Island endemic, numerically dominant predatory insect of the Hawaiian montane forest (Liebherr and Zimmerman 2000); therefore understanding the beetle’s diet could have conservation and management implications. The Hawaiian carabids are a particularly diverse fauna with 402 described endemic species descended from three colonization events: 140 Blackburnia (Platynini), 239 Mecyclothorax (Moriomorphini), and 23 Bembidion (Bembidiini) (Liebherr and Zimmerman 2000, Liebherr 2008a, Liebherr 2008b). Of the 140 Blackburnia, only 23 species in the subgenus Colpocaccus possess fully functioning flight wings. Blackburnia hawaiiensis is of interest for a diet study because it is the only Blackburnia on Hawai‘i Island with fully functioning flight wings, allowing for increased foraging capabilities (Liebherr and Zimmerman 2000).

Worldwide, carabids are commonly considered to be opportunists, consuming a wide variety of prey items such as Diptera and Lepidoptera larvae, Collembola, and aphids (Borror and White 1970, Hagley et al. 1982, Fawki and Toft 2005). Liebherr (pers. com. 2014) observed remnants of spiders, caterpillars, and fruiting bodies of moss mats in the gut contents of Blackburnia spp. Although carabids are usually assumed to be generalist predators (Allen 1979, Hagley et al. 1982, Ekbom 1992, Bilde and Toft 1997, Ball and Bousquet 2001, Fawki and Toft 2005) several are phytophagous (Honek et al. 2003, Menalled et al. 2007) and oligophagous (Hatteland et al. 2011, Brandmayr and Brandmayr 2013), such as Scaphinotus, which feed exclusively on snails (Ober et al. 2011).

Our objective was to determine the diet of B. hawaiiensis, testing the hypothesis that this beetle is a generalist predator, consuming more than one arthropod order. Ultimately, we intended to enhance basic knowledge of Hawaiian arthropods where conservation efforts are hindered by a lack of taxonomic [End Page 246] and ecological data (Howarth and Mull 1992, Medeiros et al. 2013). We tested this hypothesis by analyzing the δ15N and δ13C isotope values of these beetles and other taxa collected in the same microhabitat, and by using genetic metabarcoding on the dissected gut contents of B. hawaiiensis.

materials and methods

Site Description

Three geographically separated, replicated sites on Hawai‘i Island were chosen for this study: Nāhuku (N 19° 24′; W 155° 14′), Ka‘iholena (N 19° 11′; W 155° 35′), and Pu‘u Maka‘ala (N 19° 32′; W 155° 13′). Substrates at Nāhuku consist of Kīlauea volcanic eruptions and Ka‘iholena and Pu‘u Maka‘ala were formed from Mauna Loa eruptions. The selection of sites was guided by Light Detection and Ranging (LIDAR), ease of access, and local inspection to control for abiotic factors: elevation range of 1120–1210 m, a substrate age of 200–750 years, and a mean annual temperature of 15.5–16 °C. The three sites were also characterized as tropical montane forest with a Metrosideros polymorpha canopy, tree fern (Cibotium spp.) subcanopy, and active populations of B. hawaiiensis (Figure 1). Annual rainfall was slightly higher at one of the sites; the mean annual precipitation of Ka‘iholena and Nāhuku is 2,500–3,000 mm/yr while Pu‘u Maka‘ala is 4,500–5,700 mm/yr (Giambelluca et al. 2013). Sites were ground-truthed to select areas with low human and feral ungulate disturbance and predominantly native vegetation.

Arthropods for both molecular and stable isotope techniques were collected along fence lines for approximately 600 m in order to minimize damage to delicate native vegetation during the study. Most arthropod collections took place at dusk and dawn during several trips May–November 2014, which is the dry season when B. hawaiiensis is most prevalent and active in the forest (Liebherr pers. com. 2014). Supplemental SIA collection occurred in Summer 2015 to increase sample sizes. All arthropods were collected at ~1.3 m above ground in the lower forest arthropod community. Arthropods were collected by gently shaking brown Cibotium fronds over a beating sheet for 10 s and aspirating the arthropods into snap-cap tubes.

Arthropod Collection for Stable Isotope Analyses

Focal taxa selection for stable isotope analyses was based on the following criteria: observed in the same microhabitat as B. hawaiiensis (brown Cibotium leaves), common to all three sites, and easily identifiable in the field based on morphological characteristics. Taxa included adult Nabis (Hemiptera: Nabidae), Pagiopalus (Araneae: Philodromidae), Collembola, Leptogryllus (Orthoptera: Gryllidae), Laupala (Orthoptera: Gryllidae), and Cibotium (Cyatheales: Cibotiaceae) and Metrosideros (Myrtales: Myrtaceae) leaf litter. Specimens were sorted by taxon and kept alive for 2–4 days in order to allow their digestive tracks to clear, reducing the contamination of gut material in the samples (McNabb et al. 2001). Taxa were stored at 20 °C for a maximum of one month before further processing. To prepare the specimens for SIA, they were oven-dried at 60 °C for ≤48 hrs, pulverized, then placed in 8 × 5 mm tin capsules. Individuals were pooled until 0.5–1 mg dryweight of arthropod and 1–2 mg of plant material was sampled (typically 2 Nabis, 2 Pagiopalus, 20 Collembola, 3 Leptogryllus, 4 Laupala, and 4 Cibotium leaflets and 5 Metrosideros leaves). Isotopic compositions and δ15N and δ13C concentrations of each sample were measured using an elemental analyzer coupled to an isotope ratio mass spectrometer (EA-IRMS; PN-150 Costech® Auto Sampler and Elemental Analyzer with Finnegan Conflow III® regulator and Isodat 3.0® software; Thermo Electron Corporation, San Jose, CA, USA) by the first author at the University of Hawai‘i at Hilo Analytical Laboratory ( Peach leaf standard (NIST 1547) was also run with the samples for quality control.

Blackburnia Collection for Genetic Analyses of Gut Contents

Blackburnia hawaiiensis were transferred within 30 min after collection to 2 ml screwtop micro stand tubes containing 95% ethanol [End Page 247] (to prevent digestion of prey in the gut of the beetles) and then stored at 80 °C until further processing. Blackburnia hawaiiensis were identified and sexed under a Leica MZ-2500 binocular dissecting scope according to Liebherr and Zimmerman (2000).

Figure 1. (A) Study sites on Hawai‘i Island including substrate ages and rainfall isohyets. (B) Pu‘u Maka‘ala. (C) Nāhuku. (D) Ka‘iholena. Substrate layers were created by USGS () and precipitation layer created by Rainfall Atlas Hawai‘i (). Ka‘iholena and Pu‘u Maka‘ala were formed from a Mauna Loa volcanic eruption and Nāhuku was that of Kīlauea.
Click for larger view
View full resolution
Figure 1.

(A) Study sites on Hawai‘i Island including substrate ages and rainfall isohyets. (B) Pu‘u Maka‘ala. (C) Nāhuku. (D) Ka‘iholena. Substrate layers were created by USGS (Sherrod et al. 2007) and precipitation layer created by Rainfall Atlas Hawai‘i (Giambelluca et al. 2013). Ka‘iholena and Pu‘u Maka‘ala were formed from a Mauna Loa volcanic eruption and Nāhuku was that of Kīlauea.

Molecular Techniques

Primer design and PCR

A reference sequence for B. hawaiiensis was created by first designing Platynini carabid-specific primers using mitochondrial cytochrome-oxidase I (COI) barcode sequences obtained from GenBank for 68 Platynini species. Sequences were aligned in Geneious v. 8.1.6 (Biomatters, Aukland, NZ). Conserved regions of DNA flanking the 157-bp region of the Zeale primers (ZBJ-ArtF1c and ZBJ-ArtR2c) for gut content metabarcoding (Zeale et al. 2010, Bohmann et al. 2011) of the 68 available sequences were considered for potential primer synthesis. The Platynini primers were designed to amplify a 230-bp barcode region of carabid mitochondrial DNA (F: AGATATTGGAACWTTATATTT-TATTTTTGG, R: ATCAAAAACTTATAT-TATTTATTCGAGG). Blackburnia hawaiiensis samples were amplified using: initial denaturation at 95 °C (3 min); 30 cycles of 95 °C (30 s), 53 °C (45 s); and 72 °C (45 s), with a final extension at 72 °C (5 min). Platynini primers were tested on 3 individuals from each site, yielding identical sequences (Genbank accession No. MH733243). [End Page 248]

The Zeale primers amplify a wide range of arthropod orders (Zeale et al. 2010, Bohmann et al. 2011, Piñol et al. 2014a). Preliminary testing of these primers on various Hawaiian taxa common to the environment in which B. hawaiiensis was collected, was conducted in order to find the optimal annealing temperature for a wide range of Hawaiian arthropods including Drosophila heteroneura (Diptera: Drosophilidae), Laupala spp., Lepidoptera spp., Leptogryllus spp., Nabis spp., Eupetinus and Prosopeus spp. (Coleoptera: Nitidulidae), and Triginidium spp (Orthoptera: Trigonidiidae). Optimal annealing temperature for all taxa was determined to be 51 °C and used for further HTS.

DNA extraction

For HTS, B. hawaiiensis foregut, midgut, and hindgut were dissected, homogenized, and extracted with the Qiagen DNeasy Blood and Animal Tissue Kit (Qiagen Ltd, Crawley, UK) according to the manufacturer’s protocol. Five gut content extractions from each site were used for HTS, including 8 males and 7 females (Ka‘iholena 2 males and 3 females; Nāhuku 3 males and 2 females; Pu‘u Maka‘ala 3 males and 2 females). DNA quality was checked on a 1.5% agarose gel.

High-throughput sequencing

Samples were prepared using the Ion Torrent recommendations for preparing amplicon libraries without fragmentation. The Ion Plus Fragment Library Sequencing Kit (Life Technologies) was used with minor modifications. All 15 samples were pooled into an equal-molar concentration of 15 ng/μl, with a 20-cycle initial PCR amplification to reduce the amplification bias. Additionally, the end repair reagent volumes were reduced by 75% and the adapter ligation and nick repair reagent volumes by 50% to compensate for the low throughput of DNA yields from gut extractions.

Sample purity was evaluated (DNA High Sensitivity kit, Bioanalyzer 2100, Agilent Technologies) and quantified (Qubit dsDNA HS Assay Kit, Qubit 3.0 Fluorometer, Life Technologies) between all library steps. Prior to sequencing, the Kappa Biosystems Ion Torrent Library Quantification Kit was used to determine the ideal concentration of template needed by quantitative real-time PCR. The samples were diluted, amplified (Template Hi-Q amplification, Ion One Touch 2 system), and sequenced on the Personal Genome Machine (PGM) using the Hi-Q Kit with the Ion Torrent 318v2 chip (Life Technologies) as described by the manufacturer (Life Technologies). The sequencing chemistry for 200-bp read length and version 4.4.3 of the Torrent Suite software was used for base calling (Life Technologies). All library preparation and Ion Torrent sequencing were performed by the first author at the University of Hawai‘i at Hilo Evolutionary Core Genomics Laboratory.

Data Analysis

Analysis of stable isotopes

The stable isotopic composition of element X is expressed as a difference in ratios and in parts per thousand: δX(‰)= (Rsample/Rstandard − 1) × 1,000 where Rsample and Rstandard are the ratio of heavy to light isotope (13C:12C and 15N:14N) of the sample and standard (Craig 1953). Carbon isotope data were normalized to USGS 40 (δ13C vs VPDB = −26.4) and USGS 41 (δ13C vs VPDB 37.6) (SD = 0.2‰). Nitrogen isotope data were normalized to USGS 40 (δ15N vs Air = −4.5) and USGS 41 (δ15N vs Air −47.6) (SD = 0.2‰).

R version 3.6.3 was used for all statistical analyses (R Core Team 2020). We fit linear mixed-effects models using the lmer function provided by the lme4 package for R (Bates et al. 2015) using site as a random effect with taxa as the predictor, and δ13C and δ15N as the response variables in their respective models. Post hoc tests for differences between model coefficients were done using the mltcomp package (Hothorn et al. 2008), comparing linear combinations of δ13C and δ15N B. hawaiiensis values to all other taxa. Logically constrained multiplicity adjustments were used to control multiple comparison P-values (Westfall et al. 1999).

High-throughput sequencing bioinformatic analysis

Torrent suite software sorted all sequences by barcode into separate BAM files, [End Page 249] which were then converted into FASTQ files using a simple stream editor (sed) script. Sequences were then filtered and clustered using USEARCH (Edgar 2010) according to (Krehenwinkel et al. 2017b). BLASTn was used to assign Operational Taxonomic Unit (OTU) clusters to order (McGinnis and Madden 2004) using a reference library containing NCBI COI databases concatenated with the COI B. hawaiiensis sequence, using a minimum similarity of 95%. Only order classification information was retained because of the limits of HTS such as sequencing accuracy and the limited reference database available.


Stable Isotope Analysis

Blackburnia hawaiiensis had the highest δ15N of all taxa and similar δ13C to all arthropods (Figure 2, Table 1). For both δ15N and δ13C, the linear mixed-effects models with taxa as the predictors were preferred over the null model (Table 2, Table 3). Blackburnia hawaiiensis δ15N was not significantly different from that of Nabis (P = .13) and significantly

Figure 2. δ15N and δ13C isotope ratios of B. hawaiiensis, Pagiopalus, Nabis, Laupala, Leptogryllus, Collembola, Metrosideros, and Cibotium. Colored dots are mean ± 1 SE, and open dots are raw data points. Taxa are listed in descending δ15N values.
Click for larger view
View full resolution
Figure 2.

δ15N and δ13C isotope ratios of B. hawaiiensis, Pagiopalus, Nabis, Laupala, Leptogryllus, Collembola, Metrosideros, and Cibotium. Colored dots are mean ± 1 SE, and open dots are raw data points. Taxa are listed in descending δ15N values.

Table 1. Average (δ13C and (δ15N Isotope Values (±1 SE) of Study Taxa. Number of Samples Analyzed (N) of Each Taxa is Reported Below. Taxa are Listed in Decreasing (δ 15N from Top to Bottom
Click for larger view
View full resolution
Table 1.

Average (δ13C and (δ15N Isotope Values (±1 SE) of Study Taxa. Number of Samples Analyzed (N) of Each Taxa is Reported Below. Taxa are Listed in Decreasing (δ 15N from Top to Bottom

[End Page 250]

Table 2. Multiple Comparisons of B. hawaiiensis to All Other Taxa Output from the Linear Mixed Effects Model Using Taxa as the Predictor and (δ15N as the Response with Site as a Random Effect (δ15N ~ Taxa + − 1+ (1 | Site)). Null Deviance 364.7 on 83 df Residuals, AIC 370.7. Alternative Deviance 241.0 on 76 df residuals, AIC 261.0. Random Effects Standard Deviation of 0.404, 86 Observations on 3 Groups (Sites)
Click for larger view
View full resolution
Table 2.

Multiple Comparisons of B. hawaiiensis to All Other Taxa Output from the Linear Mixed Effects Model Using Taxa as the Predictor and (δ15N as the Response with Site as a Random Effect (δ15N ~ Taxa + − 1+ (1 | Site)). Null Deviance 364.7 on 83 df Residuals, AIC 370.7. Alternative Deviance 241.0 on 76 df residuals, AIC 261.0. Random Effects Standard Deviation of 0.404, 86 Observations on 3 Groups (Sites)

Table 3. Multiple Comparisons of B. hawaiiensis to All Other Taxa Output from the Linear Mixed Effects Model Using Taxa as the Predictor and (δ13C as the Response with Site as a Random Effect (δ13C ~ Taxa + 1+ (1 | Site)). Null Deviance 326.7 on 83 df Residuals, AIC 332.7. Alternative Deviance 206.1 on 76 df Residuals, AIC 226.1. Random Effects Standard Deviation of 0.287, 86 Observations on 3 Groups (Sites).
Click for larger view
View full resolution
Table 3.

Multiple Comparisons of B. hawaiiensis to All Other Taxa Output from the Linear Mixed Effects Model Using Taxa as the Predictor and (δ13C as the Response with Site as a Random Effect (δ13C ~ Taxa + 1+ (1 | Site)). Null Deviance 326.7 on 83 df Residuals, AIC 332.7. Alternative Deviance 206.1 on 76 df Residuals, AIC 226.1. Random Effects Standard Deviation of 0.287, 86 Observations on 3 Groups (Sites).

differed from all other taxa (P = .007 for Pagiopalus, P < .001 for all other taxa) (Table 2). Blackburnia hawaiiensis δ13C was significantly different from both plant taxa, Metrosideros and Cibotium (P < .001), but was not significantly different from all other arthropods (Table 3).

High-Throughput Sequencing

The Ion Torrent run produced 4,477,826 DNA sequencing reads with 63% loading. The mean read length was 211 bp. Each barcode returned on average 254,811 ± 5,754.22 reads per barcode with a minimum number of reads at 199,987 and a maximum of 270,275. After USEARCH filtering, we recovered an average of 29,279 ± 871.841 SE reads per specimen, ranging from 21,912 to 34,077 reads. 0.05% of the usable sequences were prey; therefore, an average of 200 predator reads was generated for each prey sequence.

Prey items were amplified in eight of the 15 gut content samples, five females, and three males. The majority of prey reads belonged to Diptera (98%, 215 reads). Lepidoptera was found only in the gut contents of a single individual (2%, 5 reads).


Blackburnia hawaiiensis Isotopic Differences

Because δ15N often indicates trophic level (Post 2002, Shiels et al. 2018), our δ15N mixed-effects model suggests B. hawaiiensis is in a higher trophic category than all tested taxa except for Nabis, or damsel bugs. These data indicate B. hawaiiensis is on a similar trophic level to that of Nabis, which are considered generalist predators that feed on a variety of other insects (Lattin 1989, Stasek et al. 2018). On average, B. hawaiiensis δ15N is more than 1.3 values higher than the other significantly different taxa (Table 1). Interestingly, δ15N of B. hawaiiensis was significantly higher than Pagiopalus, a genus of running crab spider or Philodromidae, which have also been documented to consume a variety of arthropods (Huseynov 2008). In Hawai‘i, stable isotopic studies have been conducted on native spiders, and Kennedy et al. (2018) in particular conducted research in close proximity to Pu‘u Maka‘ala. Kennedy et al.’s (2018) data suggest similar δ15N for Tetragnathid and Ariamnes spiders and B. hawaiiensis.

Signatures of δ13C tend to reflect the food source of an animal (Cabana and Rasmussen 1996, Post 2002, Hobson et al. 2010). Our linear mixed-effects model showed that [End Page 251] B. hawaiiensis δ13C values were similar to all the arthropod taxa used in this study and therefore are all potential prey items for B. hawaiiensis. Blackburnia hawaiiensis δ13C values were different from the tested plant material indicating that the beetle is unlikely to be consuming plant material. Additionally, our δ13C data suggest that Collembola is a likely base for the Hawaiian montane forest arthropod food web, and better suited than Metrosideros and Cibotium for future SIA studies using mixing models.

Future research using Bayesian mixing models such as MixSIR (Moore and Semmens 2008) or SIAR (Parnell et al. 2010) would aid in determining the trophic position and spatial diet contributions of arthropod prey. In order to use these mixing models, feeding studies and larger sample sizes are necessary. Controlled feeding experiments of B. hawaiiensis for determining the trophic enrichment factor could be possible, as Liebherr (2000) successfully raised a closely related species, although a self-perpetuating colony was not established.

High-Throughput Sequencing Analysis

Our HTS analyses suggest that B. hawaiiensis is consuming at least two orders of arthropods: Diptera and Lepidoptera. However, these orders may not reflect the full spectrum of diet due to sample size (N = 15), noting that prey items were amplified in only eight of the individuals’ gut contents. Interestingly, host amplicons were abundant despite reducing the amount of available host material through dissection and limiting the initial PCR to 20 cycles. Our results were similar to that of Piñol et al. (2014a) and Gomez-Polo et al. (2015), who both used the brute-force approach with 40 initial PCR cycles resulting in over 99% predator sequences. Interestingly, in Piñol et al. (2014b), reducing PCR cycles only increased the proportion of the less well-amplified taxa, not the presence of the taxa themselves.

High-throughput sequencing is a powerful and rapidly evolving technique but there are still many shortcomings that can make the data difficult to interpret. For example, the absence of site-specific reference databases and the potential for sequencing errors in HTS prevent the assigning of reads to species level. Blocking primers or host-specific primers modified with a C3 spacer at the 30 end of the forward universal primer can also be used to reduce the number of predator sequences (Vestheim and Jarman 2008, Deagle et al. 2013, Gomez-Polo et al. 2015). However, blocking primers have not proven to be useful for arthropod predators that consume closely related organisms (Gomez-Polo et al. 2015, Piñol et al. 2014b). For example, Piñol et al. (2014b) found that the use of blocking oligonucleotides reduced predator sequence by less than one order of magnitude. After the alignment of the COI arthropod sequences available in Genbank, it became clear that this approach could not be used for B. hawaiiensis because high conservation of the COI target sequence would have led to significant underestimation of insect orders within the samples. The COI region is highly conserved and thus significant proportions of all insect orders examined may have been inappropriately blocked by blocking primers, as seen in Toju and Baba (2018). The addition of other barcoding genes such as 16S ribosomal RNA and ITS (Krehenwinkel et al. 2019), a Hawaiian arthropod reference library, and a larger sample size may improve genetic diet analyses and perhaps identify prey items that were not detected in this study.

Blackburnia hawaiiensis, A Generalist Predator

Knowledge of feeding ecology can ultimately improve the effectiveness of land management and conservation practices for species and ecosystems. In this study, we found evidence that B. hawaiiensis is a generalist predator that consumes at least two orders of arthropods, and has a trophic position similar to that of damsel bugs. Unfortunately, Diptera and Lepidoptera larvae were not abundantly collected through the beating method and were not included as potential prey items in the SIA. Liebherr and Zimmerman (2000) suggested that the foraging activity of [End Page 252] Hawaiian carabids is not limited by spatial constraints. The beetles can be found in moss mats, under logs and rocks, and on ‘ōhi‘a trees and ‘ie‘ie vines (Pandanaceae: Freycinetia arborea) in addition to Cibotium fronds where B. hawaiiensis was collected in this study, a nod to their generalist eating habits (Liebherr and Zimmerman 2000). Indeed, larvae of both Diptera and Lepidoptera orders could be easily captured by a mobile predator in decaying woody tissue. Although further research is necessary to determine the full scope of B. hawaiiensis diet, the results from this study contribute to the limited but advancing Hawaiian arthropod diet ecology literature. The methodological approaches used in this study will be useful to document the natural history and diet of arthropods and could ultimately improve the study of trophic ecology.

author contributions

KR, CPE, and DKP conceived and designed the experiments. KR and CPE conducted field work and identification of arthropods. KR performed the experiments, statistical analysis, and wrote the manuscript. KR, CPE, and DKP revised the manuscript.

K. Roy
Tropical Conservation Biology and Environmental Science, University of Hawai‘i, Hilo, USA.
U.S. Geological Survey, Pacific Island Ecosystems Research Center, Kīlauea Field Station, Hawai‘i National Park, Hawai‘i, USA.
C. P. Ewing
Tropical Conservation Biology and Environmental Science, University of Hawai‘i, Hilo, USA.
California Department of Forestry and Fire Protection, 5800 Chiles Road, Davis, CA, USA.
D. K. Price
Tropical Conservation Biology and Environmental Science, University of Hawai‘i, Hilo, USA.
School of Life Sciences, University of Nevada, Las Vegas, 4505 S Maryland Pkwy, Las Vegas, NV 89154, USA.
This study was funded by the Hau‘oli Mau Loa Foundation. Manuscript accepted 27 May 2020.
Corresponding author (e-mail: Roy


We would like to thank Dr. Rebecca Ostertag, Thomas Fezza, Anne Veillet, Tara Holitzki, Dr. Renee Bellinger, and the Geib lab at USDA-ARS-DKI-PBARC (including Dr. Scott Geib, Brian Hall, and Theodore DeRego) for making this project possible. Dr. Henrick Krehenwinkel, Dr. Jun Y. Lim, and Dr. Susan R. Kennedy assisted in genetic data analysis and provided valuable advice. We thank the DLNR, TNC, HAVO, and DOFAW for site access and permitting. A special “mahalo” to the Hau‘oli Mau Loa Foundation for financial support. We also thank Dr. Carter T. Atkinson, Dr. Dennis Lapointe, and two anonymous reviewers for review of this article. This project was conducted for the fulfillment of MSc in Tropical Conservation Biology and Environmental Sciences at the University of Hawai‘i at Hilo. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.


This article was created by a U.S. government employee and is in the Public Domain. Public Domain information may be freely distributed and copied, but it is requested that in any subsequent use the U.S. Department of Interior U.S. Geological Survey and the journal, Pacific Science, be given appropriate acknowledgment.

Literature Cited

Allen, R. T. 1979. The occurrence and importance of ground beetles in agricultural and surrounding habitats. Pages 485–505 in T. L. Erwin, G. E. Ball, and D. R. Whitehead, eds. Carabid beetles: their evolution, natural history and classification. Dr. W. Junk Publishing, The Hague-Boston-London.
Ball, G. E., and Y. Bousquet. 2001. Carabidae Latreille. Pages 32–132 in R. H. Arnett Jr, and M. C. Thomas, eds. American beetles, Volume 1 (Archostemata, Myxophaga, Adephaga, Polyphaga: Staphylinifornia). CRC Press, Boca Raton, FL, USA.
Bates, D., M. Mächler, B. Bolker, and S. Walker. 2015. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 67:1–48.
Bilde, T., and S. Toft. 1997. Consumption by carabid beetles of three cereal aphid species relative to other prey types. Entomophaga 42:21–32.
Bohmann, K., A. Monadjem, C. Lehmkuhl Noer, M. Rasmussen, M. R. K. Zeale, E. Clare, G. Jones, E. Willerslev, M. Thomas, and P. Gilbert. 2011. Molecular dietary analysis of two African free-tailed bats (Molossidae) using high throughput sequencing. PLoS One 6:e21441.
Borror, D. J., and R. E. White. 1970. A field guide to the insects of American North of Mexico. Houghton Mifflin Company, Boston.
Brandmayr, P., and T. Zetto Brandmayr. 2013. Food and feeding behaviour of Lincinus species (Coleoptera: Carabidae Lincinini). Ital. J. Zool. 20:171–181. DOI: 10.1080/00269786.1986.10736497.
Cabana, G., and J. B. Rasmussen. 1996. Comparison of aquatic food chains using nitrogen isotopes. Proc. Natl. Acad. Sci. 93:10844–10847.
Carreon-Martinez, L., and D. D. Heath. 2010. Revolution in food web analysis and trophic ecology: diet analysis by DNA and stable isotope analysis. Mol. Ecol. 19:25–27.
Craig, H. 1953. The geochemistry of the stable isotopes of carbon. Geochim. Cosmochim. Acta 3:53–92.
Deagle, B. E., A. C. Thomas, A. K. Shaffer, A. W. Trites, and S. N. Jarman. 2013. Quantifying sequence proportions in DNA-based diet study using Ion Torrent amplicon sequencing: which counts count? Mol. Ecol. Resour. 13:620–633.
Edgar, R. C. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26:2460–2461.
Ekbom, B. 1992. Diet, fecundity and egg size in some polyphagous predatory carabid beetles. Entomol. Exp. 65:129–140.
Fawki, S., and S. Toft. 2005. Food preferences and the value of animal food for the carabid beetle Amara similata (Gyll.) (Col., Carabidae). J. Appl. Entomol. 129:551–556.
Giambelluca, T. W., Q. Chen, A. G. Frazier, J. P. Price, Y. L. Chen, P. S. Chu, J. K. Eischeid, and D. M. Delparte. 2013. Online Rainfall Atlas of Hawai‘i. B. Am. Meterol. Soc. 94:313–316. DOI: 10.1175/BAMS-D-11-00228.1. Available at
Gomez-Polo, P., O. Alomar, C. Castane, T. Aznar-Fernandez, J. G. Lundgren, J. Piñol, and N. Agusti. 2015. Understanding trophic interactions of Orius spp. (Hemiptera: Anthocoridae) in lettuce crops by molecular methods. Pest Manag. Sci. DOI: 10.1002/ps.3989.
Hagley, E. A. C., N. J. Holliday, and D. R. Barber. 1982. Laboratory studies of the food preferences of some orchard carabids (Coleoptera: Carabidae). Can. Entomol. 114:431–437.
Hatteland, B. A., W. O. C. Symondson, R. A. King, M. Skage, C. Schander, and T. Sholhoy. 2011. Molecular analysis of predation by carabid beetles (Carabidae) on the invasive Iberian slug Arion lusitanicus. Entomol. Res. 101:675–686.
Hobson, K. A., R. Barnett-Johnson, and T. Cerling. 2010. Using isoscapes to track animal migration. Pages 273–298 in J. B. West, T. E. Dawson, G. J. Bowen, and K. P. Tu, eds. Isoscapes. Springer, Dordrecht, the Netherlands.
Honek, A., Z. Martinkova, and V. Jarosik. 2003. Ground beetles (Carabidae) as seed predators. Eur. J. Entomol. 100:531–544.
Hoogendoorn, M., and G. E. Heimpel. 2001. PCR-based gut content analysis of insect predators: using ribosomal ITS-1 fragments from prey to estimate predation frequency. Mol. Ecol. 10:2059–2067.
Hothorn, T., F. Bretz, and P. Westfall. 2008. Simultaneous inference in general parametric models. Biom. J. 50:346–363.
Howarth, F. G., and W. P. Mull. 1992. Hawaiian insects and their kin. University of Hawai‘i Press, Honolulu, Hawai‘i.
Huseynov, E. F. 2008. Natural prey of the spider Tibellus macellus (Araneae, Philodromidae). Bull. Br. Arachnol. Soc. 14:206–208.
Johnston, C. A., E. E. Wilson Rankin, and D. S. Gruner. 2018. Foraging connections: patterns of prey use linked to invasive predator diet movement. PLoS ONE 13: e0201883.
Kennedy, S. R., T. E. Dawson, and R. G. Gillespie. 2018. Stable isotopes of Hawaiian spiders reflect substrate properties along a chronosequence. PeerJ. DOI: 10.7717/peerj.4527.
Kennedy, S., J. Y. Lim, J. Clavel, H. Krehenwinkel, and R. G. Gillespie. 2019. Spider webs, stable isotopes and molecular gut content analysis: multiple lines of evidence support trophic niche differentiation in a community of Hawaiian spiders. Funct. Ecol. 33:1722–1733.
Krehenwinkel, H., S. Kennedy, S. Pekar, R. G. Gillespie. 2017a. A cost efficient and simple protocol to enrich prey DNA from extractions of predatory arthropods for large-scale gut content analysis by Illumina sequencing. Methods Ecol. Evol. 8:126–134.
Krehenwinkel, H., M. Wolf, J. Y. Lim, A. J. Rominger, W. B. Simison, and R. G. Gillespie. 2017b. Estimating and mitigating amplification bias in quantitative arthropod metabarcoding. Sci.Rep. 7:17668. DOI: 10.1038/s41598-017-17333-x.
Krehenwinkel, H., A. Pomerantz, J. B. Henderson, S. R. Kennedy, J. Y. Lim, V. Swamy, J. D. Shoobridge, N. H. Patel, R. G. Gillespie, and S. Prost. 2019. Nanopore sequencing of long ribosomal DNA amplicons enables portable and simple biodiversity assessments wit high phylogenetic resolution across broad taxonomic scale. GigaScience 8:pii:giz006. DOI: 10.1093/gigascience/giz006.
Kristensen, P., T. Riis, H. E. Dylmer, E. A. Kristensen, M. Meerhoff, B. Olesen, F. Teixeira-de Mello, A. Baatrup-Pedersen, G. Cavalii, and E. Jeppesen. 2016. Baseline identification in stable-isotope studies of temperate lotic systems and implications for calculated trophic positions. Freshw. Sci. 35. DOI: 10.1086/687284.
Lattin, J. D. 1989. Bionomics of the Nabidae. Annu. Rev. Entomol. 24:282–400.
Liebherr, J. K. 2008a. Taxonomic revision of Hawaiian Bembidion Latrielle (Coleoptera: Carabidae: Bembidiini) with a discussion of their reductive and derivative evolutionary specializations. Ann. Carnegie Mus. 77:31–78.
———. 2008b. Taxonomic revision of Mecyclothorax Sharp (Coleoptera, Carabidae) of Hawaii Island: abundant genetalic variation in a nascent island radiation. Deut. Entomol. Z. 55:19–78. DOI: 10.1002/mmnd.200800004.
———. 2000. The unity of characters: ecological and morphological specialization in larvae of Hawaiian Platynine Carabidae (Coleoptera). Invertebr. Taxon. 14:931–940.
Liebherr, J. K., and E. C. Zimmerman. 2000. Insects of Hawai‘i: Vol. 16 Hawaiian Carabidae (Coleoptera). Part 1: Introduction and Tribe Platynini. University of Hawai‘i Press, Honolulu, Hawai‘i.
McGinnis, S., and T. L. Madden. 2004. BLAST: at the core of a powerful and diverse set of sequence analysis tools. Nucleic Acid. Res. 32. DOI: 10.1093/narlgkh435.
McNabb, D., M. J. Halaj, and D. H. Wise. 2001. Inferring trophic positions of generalist predators and their linkage to the detrital food web in agroecosystems: a stable isotope analysis. Pedobiologia 45:289–297.
Medeiros, M. J., J. A. Eiben, W. P. Haines, R. L. Kahololaa, C. B. A. King, P. D. Krushelnycky, K. N. Magnacca, D. Rubinoff, F. Starr, and K. Starr. 2013. The importance of insect monitoring to conservation actions in Hawai‘i. Proc. Hawaii. Entomol. Soc. 45:149–166.
Menalled, F. D., R. G. Smith, T. Dauer, and T. B. Fox. 2007. Impact of agricultural management on carabid communities and weed seed predation. Agric. Ecosyst. Environ. 118:49–54.
Mestre, L., J. Piñol, J. A. Barrientos, X. Espadler, K. Brewitt, C. Werner, and C. Platner. 2013. Trophic structure of the spider community of a Mediterranean citrus grove: a stable isotope analysis. Basic Appl. Ecol. 14(5):413–422. DOI: 10.1016/j.baae.2013.05.001.
Moore, J. W., and B. X. Semmens. 2008. Incorporating uncertainty and prior information into stable isotope mixing models. Ecol. Lett. 11:470–480.
Ober, K., B. Matthews, A. Ferrieri, and S. Kuhn. 2011. The evolution and age of populations of Scaphinotus petersi Roeschke on Arizona Sky Islands (Coleoptera, Carabidae, Cychrini). Zookeys. 147:183–197.
Parnell, A. C., R. Inger, S. Bearhop, and A. L. Jackson. 2010. Source partitioning using stable isotopes: coping with too much variation. PLoS ONE 5:e9672. DOI: 10.1371/journal.pone.0009672. PMID: 20300637.
Piñol, J., V. San Andres, E. L. Clare, G. Mir, and W. O. C. Symondson. 2014a. A pragmatic approach to the analysis of diets of generalist predators: the use of highthroughput sequencing with no blocking probes. Mol. Ecol. Resour. 14:18–26.
Piñol, J., G. Mir, P. Gomez-Polo, and N. Agusti. 2014b. Universal and blocking primer mismatches limit the use of high-throughput DNA sequencing for the quantitative metabarcoding of arthropods. Mol. Ecol. Resour. 15(4):819–830. DOI: 10.1111/1755-0998.12355.
Polis, G. A., C. A. Myers, and R. D. Holt. 1989. The ecology and evolution of intraguild predation: potential competitors that eat each other. Annu. Rev. Ecol. Syst. 20:297–300.
Post, D. M. 2002. Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology 83:703–718.
R Core Team. 2020. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL:
Sabelis, M. W. 1992. Predatory arthropods. Pages 225–264 in M. J. Crawley, ed. Natural enemies: the population biology of predators, parasites and diseases. Blackwell Scientific, Oxford.
Sherrod, D. R., J. M. Sinton, S. E. Watkins, and K. M. Brunt. 2007. Geologic map of the state of Hawaii. US Geological Survey open-file report 2007-1089. Available at
Shiels, A. B., W. P. Bukoski, and S. R. Siers. 2018. Diets of Kauai’s invasive rose-ringed parakeet (Psittacula krameri): evidence of seed predation and dispersal in a human-altered landscape. Biol. Invasions 20:1449–1457.
Shokralla, S., J. L. Spall, J. F. Gibson, and M. Hajibabaei. 2012. High-throughput sequencing technologies for environmental DNA research. Mol. Ecol. 21:1794–1805.
Stasek, D. J., J. N. Radl, and T. O. Crist. 2018. The functional response and prey preference of generalist Nabis (Hemiptera: Nabidae) predators to leafhopper prey (Hemiptera: Cicadellidae). Can. Entomol. 150:190–200.
Thomas, A. C., S. N. Jarman, K. H. Haman, A. W. Trites, and B. E. Deagle. 2014. Improving accuracy of DNA diet estimates using food tissue control materials and an evaluation of proxies for digestion bias. Mol. Ecol. 23:3706–3718.
Toju, H., and Y. G. Baba. 2018. DNA metabarcoding of spiders, insects, and springtails for exploring potential linkage between above- and below-ground food webs. Zoological Lett. 4:4.
Vestheim, H., and S. N. Jarman. 2008. Blocking primers to enhance PCR amplification of rare sequences in mixed samples—a case study on prey DNA in Antarctic krill stomachs. Front. Zool. 5:12.
Westfall, P. H., R. D. Tobias, D. Rom, R. D. Wolfinger, and Y. Hochberg. 1999. Multiple comparisons and multiple tests using the SAS system. SAS Institute Inc., Cary, NC.
Whitaker, M. R. L., C. C. M. Baker, S. M. Salzman, D. J. Martins, and N. E. Pierce. 2019. Combining stable isotope analysis with DNA metabarcoding improves inferences of trophic ecology. PLoS ONE 14 (7):e0219070. DOI: 10.1371/journal.pone.0219070.
Zeale, M. R. K., R. K. Butlin, G. L. A. Barker, D. C. Lees, and G. Jones. 2010. Taxonspecific PCR for DNA barcoding arthropod prey in bat feces. Mol. Ecol. Resour. 11:236–244.

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.