1. Ecology
  2. Physics of Living Systems

Extreme suction attachment performance from specialised insects living in mountain streams (Diptera: Blephariceridae)

  1. Victor Kang  Is a corresponding author
  2. Robin T White
  3. Simon Chen
  4. Walter Federle
  1. Department of Zoology, University of Cambridge, United Kingdom
  2. Carl Zeiss Research Microscopy Solutions, United Kingdom
https://doi.org/10.7554/eLife.63250
Share this article
Cite this article
  1. Victor Kang
  2. Robin T White
  3. Simon Chen
  4. Walter Federle
(2021)
Extreme suction attachment performance from specialised insects living in mountain streams (Diptera: Blephariceridae)
eLife 10:e63250.
https://doi.org/10.7554/eLife.63250
  2. Download BibTeX
  3. Download .RIS

Abstract

Suction is widely used by animals for strong controllable underwater adhesion but is less well understood than adhesion of terrestrial climbing animals. Here we investigate the attachment of aquatic insect larvae (Blephariceridae), which cling to rocks in torrential streams using the only known muscle-actuated suction organs in insects. We measured their attachment forces on well-defined rough substrates and found that their adhesion was less reduced by micro-roughness than that of terrestrial climbing insects. In vivo visualisation of the suction organs in contact with microstructured substrates revealed that they can mould around large asperities to form a seal. We have shown that the ventral surface of the suction disc is covered by dense arrays of microtrichia, which are stiff spine-like cuticular structures that only make tip contact. Our results demonstrate the impressive performance and versatility of blepharicerid suction organs and highlight their potential as a study system to explore biological suction mechanisms.

eLife digest

Suction cups are widely used to attach objects to surfaces in bathrooms and kitchens. They work well on tiles and other smooth surfaces, but do not stick well to rougher materials like brick or wood because they are unable to form an air-tight seal.

Researchers have been searching for ways to improve these cups by studying how octopuses, remora fish and other sea animals use muscle-powered suction organs to stick to wet and rough surfaces. However, the experiments needed to understand the detailed mechanics of suction organs are difficult to perform on living specimens of these animals.

The aquatic larvae of a family of insects known as the net-winged midges also have suction organs that are powered by muscles. These insects survive in fast flowing mountain streams where they use their suction organs to stick to rocks underwater. However, it remained unclear how these suction organs work.

Here, Kang et al. found that net-winged midge larvae attach extremely well to a variety of surfaces. The larvae were able to withstand forces over one thousand times their body weight when attached to smooth surfaces. Even on rough materials, where human-made suction cups attach poorly, the larvae were able to withstand forces up to 240-times their body weight.

Further experiments using several microscopy approaches revealed that the suction organs of the larvae are covered in multiple spine-like structures called microtrichia that interlock with bumps and dips on a surface to help the organ remain in place. Similar structures have previously been found on the suction organs of remora fish, but are not as tightly packed together.

These findings demonstrate that net-winged midge larvae may be useful model systems to study how natural suction organs operate. Furthermore, they provide a new source of inspiration for scientists and engineers to design and manufacture suction cups capable of attaching to a wider variety of surfaces.

Introduction

Of the approximately one million known species of insects, only 325 attach using muscle-controlled suction organs (Stork, 2018; Roskov et al., 2020). These species belong to a single dipteran family, the Blephariceridae, and their larvae and pupae develop on rocks in torrential alpine streams where flow rates can exceed 3 ms–1 (Frutiger and Buergisser, 2002; Zwick, 2004; Figure 1a & b; Video 1). Each blepharicerid larva has six ventral suction organs to attach to biofilm-covered rock surfaces, where it feeds on diatoms. Using its suction organs, the larva can locomote relatively quickly and possibly over long distances: blepharicerid larvae migrate from one stone to another to find the swiftest regions of the stream (Frutiger, 1998; Mannheims, 1935). Once development is complete, the winged adult emerges from its pupa, floats to the water surface, and immediately flies away to mate and lay eggs to begin the cycle anew (Oosterbroek and Courtney, 1995; Craig, 1966).

Figure 1
Download asset Open asset
Overview of Hapalothrix lugubris and their suction attachment organs.

(a) Hapalothrix lugubris larvae live attached to rocks in torrential alpine waterways. Blue arrows indicate stream flow direction. Arrowheads highlight two larvae revealed from a brief obstruction of the waterflow. (b) H. lugubris larva (dorsal view) on natural substrate. (c) Ventral view of a larva showing its six suction organs (one organ marked by arrowhead). (d) Scanning electron micrograph showing two suction organs (arrowheads). (e-i) Computed microtomography rendering of one whole organ. A: anterior, P: posterior, D: dorsal, V: ventral. (e-ii) Side view after digital dissection showing the following structures: outer radial beams (rb), palisade layer (pl), piston cone (p), and piston muscles (pm). The cuff wall (cw) encircles the suction cavity, and the outer fringe layer (fl) encircles the disc. (f) Frontal view showing the V-notch (vn) and its pair of apodemes (vna) extending dorsally into the body. Outer cuticle has been digitally dissected to reveal the radial beams. Note the pair of piston muscles extending dorsally. V: ventral, D: dorsal. Scale bars: (c) and (d) 500 µm; (e-i), (e-ii), and (f) 100 µm.

Video 1
Download asset
Hapalothrix lugubris larvae live on rocks in torrential alpine streams.

Temporarily diverting the flow of water reveals two larvae firmly attached to the rock.

The remarkable morphology of blepharicerid suction organs is well described (Mannheims, 1935; Rietschel, 1961; Kang et al., 2019; Komárek and Wimmer, 1922). The organ superficially resembles a synthetic piston pump, with a suction disc that interacts with the surface and creates a seal, a central piston, and powerful piston muscles to manipulate the pressure, and a suction chamber with a thick cuticular wall to withstand low pressures during attachments. There are spine-like microstructures called microtrichia on the suction disc that contact glass surfaces and may increase resistance to shear forces. In addition, we have shown that a dedicated active detachment system allows the larva to rapidly detach its suction organ during locomotion (Kang et al., 2019).

While much is known about their morphology, the mechanisms involved in blepharicerid suction attachment are less well understood. Two studies to date have measured the attachment performance of blepharicerid larvae (Frutiger, 2002; Liu et al., 2020), yet neither of them offers mechanistic insights into how their suction organs cope with different surface conditions to generate strong underwater attachments.

Suction is one of the main strategies for strong and controllable underwater adhesion. Biological suction organs can adhere with high strength, rapid controllability, and reusability on smooth, rough, and biofilm-covered surfaces (Ditsche and Summers, 2014b). This is in remarkable contrast to artificial suction devices widely used in technical applications, which allow only slow control and are limited to clean and smooth surfaces. Despite their potential for bio-inspiration, there are only a few well-studied animals (namely, remora fish, clingfish, octopus, and leeches), for which the function of specific structures in biological suction attachments has been experimentally demonstrated (Beckert et al., 2015; Fulcher and Motta, 2006; Arita, 1967; Wainwright et al., 2013; Kampowski et al., 2016; Kampowski et al., 2020; Ditsche et al., 2014c; Kier and Smith, 1990; Kier and Smith, 2002; Smith, 1996). To date, mechanistic studies on biological adhesion have focused primarily on terrestrial climbing animals such as geckos, tree frogs, insects, and spiders (Lengerer and Ladurner, 2018; Federle and Labonte, 2019), and have greatly expanded our knowledge on how to achieve and control adhesion in air. Likewise, mechanistic studies on biological suction are needed to identify new strategies for generating and controlling underwater adhesion in different surface conditions.

Here we have investigated the mechanisms underlying suction attachments of blepharicerid larvae. We first conducted a detailed morphological study of Hapalothrix lugubris (Blephariceridae) to provide new insights into structures that are relevant for suction attachments. To understand how well blepharicerid larvae attach to different surfaces, we quantified their performance on smooth, micro-rough, and coarse-rough surfaces. We compared blepharicerid suction performance with that of a model terrestrial insect to investigate how two fundamentally distinct adhesive systems cope with surface roughness. Finally, we examined the function of spine-like microtrichia through in vivo visualisation of the contact zone during attachments on smooth and microstructured substrates.

Results

Morphology of the suction attachment organ of Hapalothrix lugubris

H. lugubris larvae have six ventromedian suction organs, with each organ comprising a suction disc, a central opening and a piston, a suction chamber surrounded by a thick-walled cuticular cuff, and a V-notch (Figure 1c-f and Video 2). The suction disc contacts the surface for attachment, and the piston and underlying piston muscles (Figure 1d & e) actively lower the pressure inside the suction chamber. Two apodemes attaching to the V-notch in H. lugubris mediate its muscle-controlled opening for rapid detachment of the suction organ (Figure 1f; see also Kang et al., 2019).

Video 2
Download asset
Three-dimensional rendering of a Hapalothrix lugubris suction organ based on computed microtomography data.

The video begins with a side view of the organ and its internal structures (see Figure 1c-e). Digital dissections and rendering were made using Drishti (Limaye and Stock, 2012).

The ventral disc surface of H. lugubris is covered in a dense array of microtrichia (Figure 2a-e). The suction disc-sealing rim, which seals the disc for suction attachment, closely resembles that of Liponeura cinerascens (Kang et al., 2019) and comprises a dense array of upright rim microtrichia (Figure 2b). This is different from Liponeura cordata, which has a distinct rim made up of a single row of horizontally flat rim microtrichia (Kang et al., 2019). Going from the rim to the centre of the disc, the short rim microtrichia transition into longer spine-like microtrichia (6.7 ± 0.5 µm in length and 0.56 ± 0.01 µm in mid-length diameter; mean of means ± standard error of the mean; measured from scanning electron microscopy (SEM) images of n = 2 individuals), and then again to shorter microtrichia in the centre.

Figure 2
Download asset Open asset
Ultrastructure of the suction disc.

(a-i) Scanning electron micrograph showing ventral view of the suction disc. The piston is withdrawn into the suction chamber (see Figure 1d). (a-ii) Schematic of the suction disc used for subsequent panels. Note the radial beams (rb) are beneath the outer cuticle layer and not visible in (a-i) but shown in (b) and (c). (b) Freeze-fractured suction disc (radial fracture plane; see dotted red line on suction disc schematic). The sealing rim and its short rim microtrichia are marked by an arrowhead. Internal radial beams (encircled in blue) originate from the palisade layer (magenta*). The fan-fibre space is encircled in green. C: disc centre. (c) Fan-fibres extend to the radial beams, which alternate between thin (blue) and wide beams (red). (d) Each microtrichium connects to an internal fibre; these fibres represent the ends of thicker branched fibres originating from the radial beams. Note: spine-like microtrichia point towards the disc centre (C). (e) Microtrichia are largely solid cuticular structures (arrowhead), each connected to a fan-fibre (*). (f) In vivo light microscopy shows the radial beams and the palisade layer (pl). (g) Computed microtomography (micro-CT) also shows that radial beams originate from the dorsoventral palisade layer. Centre-to-centre spacing of the beams is around 4 µm or 1.3°. vn: V-notch. Scale bars: (a) 200 µm; (b) 10 µm; (c) 6 µm; (d) 5 µm; (e) 2 µm; (f) 20 µm; (g) 40 µm.

The following imaging techniques were used to gain insights into the ultrastructure and internal organisation of the suction disc: freeze-fracture SEM, 3D models using computed microtomography (micro-CT) data, and in vivo transmitted light microscopy (Figure 2). While internal fan-fibre networks underneath the outer regions of the suction disc have been mentioned previously (Rietschel, 1961), we discovered that each internal fibre leads to a single microtrichium (Figure 2b-d). Moreover, all the microtrichia that were fractured during sample preparation appeared to be solid (in-filled) cuticular structures (Figure 2e). The small internal fibres leading into the microtrichia branch out from thicker trunks originating from the ventral side of the outer radial beams (Figure 2b-e). Another notable ultrastructural feature is the radial beams, which are solid cuticular structures that alternate between a wide and a narrow beam (Figure 2c). The beams originate from the palisades, a radial zone consisting of dorsoventral cuticular rods (Figure 2f & g). There are 72 radial beams in a 90° segment of the disc, corresponding to 288 beams per disc (assuming no interruption from the V-notch) and a centre-to-centre spacing of around 4 µm or 1.3°.

Attachment performance of blepharicerid larvae on different substrates

Effect of surface roughness on the attachment performance of blepharicerid larvae

H. lugubris attachment forces were measured using a centrifuge force tester on smooth, micro-rough, and coarse-rough surfaces (surface profiles shown in Table 1). Each specimen was wetted with a droplet of water prior to centrifugation (see ‘Materials and methods’ section for details). Interference reflection microscopy (IRM, see below) observations showed that the contact of the suction organs under these conditions was completely wet, and no air bubbles were present in the contact zone. Peak shear and normal (adhesive) forces per body weight were measured on horizontal and vertical substrates, respectively (Figure 3a & b). The test substrate had a significant effect on the peak shear force per body weight, with the larvae attaching best on smooth, followed by micro-rough, then coarse-rough substrates (Kruskal-Wallis rank sum test, χ22 = 26.3, p<0.001; p<0.05 for all pair-wise comparisons using Dunn’s post hoc tests with Bonferroni-Holm corrections). The same effect was observed for the peak normal force per body weight (Kruskal-Wallis rank sum test, χ22 = 24.2, p<0.001; p<0.05 for all pair-wise comparisons, see above).

Table 1
Surface profilometry of test substrates used to assess attachment performance.
Surface characteristics (mean ± SD)
Test surfacesRa (µm)Rq (µm)PV (µm)
Rough surfaces
Micro-rough(0.05 µm grain size)0.32 ± 0.010.40 ± 0.014.56 ± 0.22
Coarse-rough(30 µm grain size)7.97 ± 0.0610.37 ± 0.0878.82 ± 1.38
Microtextured substrates
10 × 10 µmNANA2.04 ± 0.18
3 × 3 µmNANA4.48 ± 0.08
3 × 3 µmNANA1.69 ± 0.03

  1. Ra: average roughness (mean height deviation); Rq: root-mean-squared roughness; PV: maximum peak-to-valley height; NA: not applicable.

Figure 3
Download asset Open asset
Attachment performance of Hapalothrix lugubris larvae on surfaces of varying roughness.

Hapalothrix lugubris larvae performance in (a) peak shear force per body weight and (b) peak normal adhesion force per body weight. The rotation of the centrifuge is indicated by the red circular arrow. Centre lines, boxes, whiskers, and filled dots represent the median, the inter-quartile range (IQR), 1.5 times IQR, and outliers, respectively.

Figure 3—source data 1

Data for Figure 3.

https://cdn.elifesciences.org/articles/63250/elife-63250-fig3-data1-v1.csv

Attachment performance of three blepharicerid species on smooth surfaces

The peak shear force per body weight on smooth surfaces measured for larvae from three blepharicerid species—L. cordata, L. cinerascens, and H. lugubris—was 585 ± 330, 324 ± 153, and 1120 ± 282, respectively (mean ± SD; Figure 4a). The highest overall shear force per body weight (1430) was obtained from a H. lugubris larva, while L. cordata produced a highest overall shear force of 54 mN.

Figure 4
Download asset Open asset
Attachment performance of three species of blepharicerid larvae (Hapalothrix lugubris, Liponeura cordata, Liponeura cinerascens) on smooth horizontal surface.

(a) Peak shear force per body weight. (b) Peak shear force.

Figure 4—source data 1

Data for Figure 4.

https://cdn.elifesciences.org/articles/63250/elife-63250-fig4-data1-v1.csv

Estimates of peak shear stress on smooth surfaces

Suction disc areas measured for L. cordata and H. lugubris were used to estimate the peak shear stress for blepharicerid suction attachments. Shear stresses were 41.2 ± 21.4 kPa and 39.3 ± 10.6 kPa (mean ± SD) for L. cordata and H. lugubris, respectively (Table 2). These values, however, are conservative estimates because (1) the contact area measurements included the outer fringe, which lies outside the suction disc seal and (2) we assumed that all six suction organs were in contact immediately before detachment. We thus derived more realistic estimates of the shear stresses by first correcting for the fact that the outer fringe layer amounted to 33% of the total imaged contact area (n = 18 suction discs from six individuals), and second, as larvae attach with fewer suction organs prior to detachment (Frutiger, 2002), we assumed that three organs were in contact. Factoring in these assumptions, the shear stresses were 111 ± 57.5 kPa and 117 ± 31.4 kPa (mean ± SD) for L. cordata and H. lugubris, respectively, and the normal stresses were 120.2 ± 81.9 kPa and 71.2 ± 22.2 kPa (Table 2).

Table 2
Shear and normal stress estimates for suction-based attachments of Hapalothrix lugubris and Liponeura cordata.
SpeciesShear stress (kPa)Normal stress (kPa)
Conservative(mean ± SD)Conservative(mean ± SD)nConservative(mean ± SD)Realistic(mean ± SD)n
Liponeura cordata41.2 ± 21.4111 ± 57.51040.5 ± 27.6120.2 ± 81.911
Hapalothrix lugubris39.3 ± 10.6117 ± 31.41540.5 ± 27.671.2 ± 22.210

  1. Conservative: contact area based on suction disc, inclusive of the outer fringe layer and all six organs in contact prior to detachment.

  2. Realistic: based on three organs in contact immediately prior to detachment and contact areas excluding the outer fringe layer.

Table 2—source data 1

Data for Table 2.

https://cdn.elifesciences.org/articles/63250/elife-63250-table2-data1-v1.csv

Shear attachment performance of suction organs on rough substrates compared to smooth adhesive pads

While blepharicerid attachment performance decreased with increasing surface roughness, we observed a different pattern with stick insects (Carausius morosu), which are a model terrestrial climbing insect (Figure 5). Stick insects rely on a combination of smooth adhesive pads and claws for attachment, where the former facilitates strong adhesion on smooth surfaces and the latter on coarse-rough substrates. Accordingly, we found that stick insects attached equally well to smooth and coarse-rough surfaces (one-way analysis of variance (ANOVA), F2,27 = 77.0, p = 0.97 using Tukey’s post hoc test; all tests with stick insects were conducted on dry substrates). On micro-rough surfaces, however, where neither the smooth pad nor the claws proved effective, their shear force per body weight decreased 16-fold (based on the mean of back-transformed values) compared to the smooth surface (same ANOVA as above, p<0.001 using Tukey’s post hoc test). The attachment performance of H. lugubris larvae was also affected by micro-roughness, but to a much lesser degree than in stick insects, with a twofold decrease (based on the mean of back-transformed values) in shear force per body weight. It is important to mention that blepharicerid larvae do not possess any claw-like appendages that can be used to increase grip.

Figure 5
Download asset Open asset
Comparison of shear attachment performance of Hapalothrix lugubris larvae versus stick insects (Carausius morosus) on smooth and rough surfaces.

Hapalothrix lugubris larvae attach using suction organs, whereas stick insects rely on smooth adhesive pads and claws. Sample sizes are shown with H. lugubris on the left and stick insects on the right.

Figure 5—source data 1

Data for Figure 5.

https://cdn.elifesciences.org/articles/63250/elife-63250-fig5-data1-v1.csv

In vivo visualisation of suction organs attaching to smooth and transparent microstructured substrates

The contact behaviour of H. lugubris suction discs was visualised in vivo using IRM. The attachment-detachment behaviour on smooth glass resembled closely that of the related Liponeura species (Kang et al., 2019; Frutiger, 2002): the suction disc came into close contact with the surface at the outer fringe layer, disc rim, microtrichia zone, and around the central opening (Figure 6ai, iii and Video 3). When the piston was raised away from the surface, the greater portion of the suction disc came into close contact as a result of the reduced hydrostatic pressure. When the organ was attached, the microtrichia made tip contact with the surface. No side contact was observed even when the suction disc was pulled closer to the surface as a result of the piston being raised. Detachment of the suction organ and forward movement was often preceded by an active opening of the V-notch (Video 3).

Figure 6
Download asset Open asset
In vivo visualisation of Hapalothrix lugubris suction disc contact on different substrates.

(a-i) Schematic of the microtrichia on smooth and 10 × 10 × 2 µm microstructured surface (ridges and grooves, 10 µm in width, and grooves, 2 µm deep) (shown to scale). (a-ii) On smooth glass, microtrichia made tip contact (seen as black dots under interference reflection microscopy [IRM]). Outer fringe layer (fl) is outside the seal (arrowheads). (a-iii) On 10 × 10 × 2 µm substrates, contact from microtrichia and fl was similar to the contact on the smooth surface. (b-i and b-ii) On 3 × 3 × 2 µm substrates, the microtrichia made tip contact on the ridges, as well as in the grooves, as seen in (b-iii). However, fewer microtrichia made contact within the narrow grooves compared to the 10 × 10 × 2 µm surface. Note that (b-ii) and (b-iii) differ only in the focus height. (c-i to c-iii) On 3 × 3 × 4 µm substrates, microtrichia made close contact on the ridges, but inside the deep grooves there was no contact. (d-i) Schematic of microtrichia coming into tip contact on smooth glass (inset: contact area observed under IRM). (d-ii) Schematic representing a hypothetical scenario that was not observed where the microtrichia tips bend and make side contacts. Scale bars: 3 µm for all microscopy images.

Video 3
Download asset
Suction organ of a Hapalothrix lugubris larva in action, filmed using in vivo interference reflection microscopy and a custom flow chamber.

Note the V-notch opens immediately prior to detachment.

Although the outer fringe layer and the microtrichia made close contact on smooth substrates, the outcomes were different on transparent microstructured substrates made of epoxy (Zhou et al., 2014). On the 10 × 10 × 2 µm and 3 × 3 × 2 µm substrates (ridge width × groove width × ridge height), the microtrichia made contact on both the ridges and the grooves, which are visible as black dots in the IRM recordings (Figure 6a-iii, b-i – iii, d-i). Similar to our observations on smooth glass surfaces, only the tips of the microtrichia made contact on top of the ridges and inside the grooves. Side contact of the microtrichia was not observed under IRM (see Figure 6 for a schematic of a hypothetical side contact). The outer fringe layer also made contact, although not uniformly. In contrast, on the 3 × 3 × 4 µm substrate, the microtrichia and the outer fringe layer made contact only on the ridges but not inside the grooves (Figure 6c–ii & iii). Moreover, we observed microbial organisms freely floating and moving within the 4-µm deep grooves (confirming the lack of close contact) but not on the ridges where the microtrichia and fringe layer were close to the surface. In contrast, no particulate or microbial movement was observed during the trials with the other microstructured surfaces (3 × 3 × 2 µm and 10 × 10 × 2 µm).

Discussion

Blepharicerid larvae attach with extreme strength on diverse surfaces

Blepharicerid larvae possess some of the most powerful (in terms of body weight) and complex suction organs among animals. The three species of blepharicerid larvae studied here (H. lugubris, L. cordata, and L. cinerascens) produced extreme shear forces on smooth surfaces with averages that ranged from 320 to 1120 times their own body weight. In terms of weight-specific attachment performance, the larvae performed better than all terrestrial insects measured using comparable methods (ie, whole-animal detachment experiments) (Federle et al., 2000; Grohmann et al., 2014). For example, the weight-specific shear attachment of blepharicerid larvae on smooth surfaces was 3–11 times greater than that of stick insects measured in this study. To achieve this extreme shear attachment, blepharicerid suction organs must come in close contact by generating an effective seal. Based on our in vivo visualisations of H. lugubris attaching to smooth glass underwater, the microtrichia make close contact with the surface, helping to both seal the organ and generate friction. This corroborates our previous findings on the suction disc contact behaviour with L. cinerascens and L. cordata (Kang et al., 2019). Likewise, the soft adhesive pads of stick insects make close contact on smooth surfaces, and while the weight-specific attachment forces are not as high as in blepharicerid larvae, they can withstand forces close to 100 times their body weight.

In contrast, the attachment of stick insects on micro-rough surfaces is significantly different to that of blepharicerid larvae: for stick insects, there was a 16-fold decrease in performance compared to smooth substrates, while for blepharicerid larvae, the decrease was only twofold. This difference in the impact of micro-roughness can be attributed to the two fundamentally different mechanisms of attachment: on micro-rough surfaces, neither the soft adhesive pads nor the tarsal claws of stick insects function properly (Bullock and Federle, 2011). This is in part due to the reduced effective contact area (the adhesive pads cannot mould sufficiently to the asperities) and also due to the reduced friction from tarsal claws (the claws cannot interlock with the small asperities). On the other hand, blepharicerid suction organs are still able to seal on micro-rough surfaces and microtrichia can interact with the asperities, which likely explains why their performance was not as diminished as in the stick insects. In addition, blepharicerid suction organs may adhere better to micro-rough surfaces because during partial contact, the gaps between the detached regions and the substrate are filled with water, whereas detached regions of stick insect pads are filled with air. As water is effectively incompressible and approximately 50 times more viscous than air, the water-filled contact zone can provide a much stronger resistance against detachment even under conditions of partial contact as on micro-rough substrates.

While blepharicerid larvae attached more strongly than stick insects on micro-rough surfaces, the opposite was found on coarse-rough surfaces: for stick insects, there was no difference in performance between coarse-rough and smooth surfaces, whereas blepharicerid larvae attachment decreased 11-fold. It is likely that both blepharicerid suction organs and stick insect adhesive pads are unable to cope with coarse surface roughness. The adhesive pads of both insects may be unable to fully mould to the large asperities, and the length of the microtrichia may be insufficient to reach the lower regions of the surface profile (Figure 6). Stick insects, however, have large pretarsal claws that can interlock with large asperities for strong attachment. Previous studies on dock beetles (Gastrophysa viridula) and stick insects (C. morosus) have reported that both beetle and stick insect attachments on coarse-rough surfaces decrease significantly when the claws are removed (Bullock and Federle, 2011; Scholz et al., 2010). This means that, although stick insects and dock beetles use two distinct adhesive systems (smooth versus hairy pads), the combination of the claws and the adhesive pads produces the same trend: both insects attach strongly to smooth and coarse-rough surfaces but poorly to micro-rough surfaces. In contrast, blepharicerid larvae do not have claw-like appendages and rely on suction organs for attachment. Consequently, these aquatic larvae do not follow the same trend as terrestrial climbing insects and perform the worst on coarse-rough substrates. A similar result was reported by Liu et al., 2020 using Blepharicera sp., where the larval attachment performance decreased with increasing surface roughness, although no quantitative information on attachment forces can be extracted from their study (the study also used a centrifuge, but only reported the rotation speed but not the insects’ mass and position; it is also unclear whether the larvae were wetted prior to the tests).

Blepharicerid suction organs withstand stresses comparable to those by other biological suction organs

Although attachment forces per body weight help to assess the performance from a biological perspective, the attachment force per contact area, or the stress, is needed to make comparisons between animals that adhere to surfaces. In blepharicerid larvae, the shear stress on smooth substrates ranged from 39 to 117 kPa, and from 24 to 120 kPa for normal stress (the range in values arises from the assumptions used to calculate the contact area and the number of organs remaining in contact immediately prior to detachment). These values are similar to those reported in the literature for suction attachments of other animals on smooth substrates: the remora fish can withstand 93 kPa in shear and 38 kPa in the normal direction (Fulcher and Motta, 2006); octopus can resist normal stresses of up to 271 kPa, squids up to 830 kPa, and lumpsucker fish up to 102 kPa (Smith, 1996; Davenport and Thorsteinsson, 1990). Like the cephalopods, blepharicerid suction stress can surpass 101 kPa (1 atm at standard sea level and temperature), with one L. cordata reaching 228 kPa (realistic estimate; 77 kPa if based on conservative assumptions, as outlined in ‘Results’ section). Although a pressure difference of 1 atm is considered the upper threshold for suction attachments in air, this is not the case if the contact zone is completely wet and bubble-free, as the strong cohesion of water allows suction stresses to exceed 1 atm (Smith, 1991; Smith, 1996; Wang et al., 2019; Wang et al., 2020). Even if they do not reach 1 atm, blepharicerid larvae can generate sufficient attachment force to resist the fast flow rates in their natural habitat.

Ultrastructural components may help to stabilise the suction disc under high stress

Recent work on cupped microstructures (which resemble microscopic suction cups) has revealed the mechanisms of failure in underwater suction attachments: (1) under sustained tensile stress, the rim slides inwards and the rim diameter contracts by ~30%; (2) immediately prior to detachment, sections of the rim buckle inwards, leading to adhesive failure (Wang et al., 2019). Similar failure modes were also reported for macroscopic suction cups (Ditsche and Summers, 2019). Two structural features of the blepharicerid suction organs could represent adaptations to counter the aforementioned failure mechanisms seen in cupped microstructures. First, the internal radial beams can provide structural support to reduce inward sliding and buckling of the suction disc rim. Similar to how the flexible membrane of an umbrella is stiffened by radial spokes, these stiff cuticular radial beams can stabilise the suction disc when the organ is under high tensile stress. Bones within the clingfish suction organ may also prevent inward sliding (Wang et al., 2020). While we have yet to visualise blepharicerid suction organs fail under extreme forces, their powerful attachments suggest that they possess mechanisms to counter suction cup failure.

The second morphological feature that may reduce inward sliding of the rim is the numerous microtrichia in the contact zone of the blepharicerid suction disc. The microtrichia can interlock with surface asperities and minimise inward sliding. This has been reported for the remora fish, which have stiff posterior-facing structures called spinules within their suction pads that passively engage with asperities during high-drag conditions (Beckert et al., 2015; Fulcher and Motta, 2006), as well as for the clingfish, where a hierarchical system of rods and filaments near the periphery of the suction organ may increase friction on rough surfaces (Ditsche and Summers, 2019). Similarly, blepharicerid microtrichia are naturally angled (~40° to 50° relative to the horizontal) and point towards the centre of the suction disc (Kang et al., 2019). Hence, inward sliding would passively lead to additional interlocking of the microtrichia tips with rough surfaces, thereby loading the microtrichia along their axis.

To interlock effectively, structures like the remora spinules need to be stiff and strong (Dai et al., 2002; Wang et al., 2017). The blepharicerid microtrichia are indeed likely to be stiff structures since (1) they are made of solid cuticles (a composite material of chitin fibres embedded in a matrix of proteins), and the dense, sclerotised cuticle can reach high elastic moduli [Parle et al., 2017]; (2) we never observed any microtrichia in side contacts, even on micro-structured surfaces when the disc was pressed into contact. From the observation that microtrichia only make tip contact, and assuming that they are loaded equally, we can give a conservative lower estimate for the elastic modulus of the microtrichia cuticle by following Goss and Chaouki, 2016 and modelling the microtrichium as a cylindrical beam loaded at an angle (Figure 7; see ‘Materials and methods’ section for details). Based on this model, the microtrichia cuticle must have a stiffness of at least 0.3–0.4 GPa, similar to the stiffness of wood and bone (Vincent and Wegst, 2004), to prevent any side contact (for an angle to the disc surface of 40° to 50°; see Figure 7). It is not unlikely that the microtrichia cuticle is even stiffer: sclerotised cuticles can have an elastic modulus of up to 20 GPa (Vincent and Wegst, 2004; Wegst and Ashby, 2004; Sykes et al., 2019). This supports our idea that microtrichia can maintain tip contact during interactions with rough surfaces and could serve a similar function to stiff remora spinules.

Figure 7
Download asset Open asset
Estimate of the elastic modulus of the microtrichia cuticle needed to maintain tip contact during attachment.

A microtrichium is modelled as a cylinder (length L and diameter D: 6.7 µm and 0.56 µm, respectively) that is loaded with a peak normal force (FN) and that makes tip contact at an angle (40° to 50°). See text for further details on the assumptions and the model.

It is worth mentioning that, while stiff interlocking structures play an important role in increasing friction on rough surfaces, an effective seal is crucial for attachments on both smooth and rough surfaces. In remora fish and clingfish, a soft rim helps the organ to mould to the surface and form a seal (Wang et al., 2020). In bio-inspired suction devices as well, it is the combination of stiff structures and a soft sealing rim that generates the strongest attachments to smooth and rough surfaces (Ditsche and Summers, 2019; Wang et al., 2017). The detailed biomechanics of how the dense array of microtrichia produces a tight seal on rough surfaces and interlocks with small substrate asperities is beyond the scope of this study and remains to be explored in future work.

Blepharicerid suction organs are unique among insects and specialised for fast-flow conditions

We have demonstrated that blepharicerid suction organs can attach with extreme strength to both smooth and rough surfaces. Despite the potential strength of each attachment, the larvae are surprisingly mobile in their natural habitat (Frutiger, 1998; Federle and Labonte, 2019). Both attachment force and mobility are required for blepharicerid larvae to survive in their challenging habitats, which include raging alpine torrents and areas near the base of waterfalls. Currently, blepharicerid suction organs are the only examples of piston-driven suction organs in insects. While the circular setae of male diving beetles (Dytiscidae) are also considered to be suction organs, the two systems have markedly different morphologies: (1) the ventral surfaces of circular setae are comparatively smooth; (2) circular setae lack a muscle-driven central piston; (3) there are no known mechanisms for rapid detachment in dytiscid circular setae (Chen et al., 2014; Karlsson Green et al., 2013; Nachtigall, 1974). As male dytiscid beetles use their suction organs to attach to the smooth sections of the female’s pronotum and elytra, their attachment works best on smooth surfaces and their performance declines strongly on rough surfaces (Karlsson Green et al., 2013). Researchers have suggested that male and female dytiscid beetles are engaged in an evolutionary arms race driven by sexual conflict, as female cuticular surfaces are modified to hinder male attachment with suction discs (Karlsson Green et al., 2013; Bergsten et al., 2001). Interestingly, it appears that male beetles did not evolve friction-enhancing structures on their circular setae to facilitate adhesion to rougher regions of the female cuticle. In contrast, blepharicerid suction discs are densely covered in microtrichia that likely enhance the grip on rough surfaces in high-drag conditions. This difference in morphology may be based on function or phylogenetic constraints: male diving beetles may not need to attach to rough elytra if their setae can generate sufficient attachment forces on smooth cuticle alone; alternatively, beetle setae may be more limited in the structures that can be developed from them, compared to the blepharicerid organ, which is highly complex and multicellular.

A more comparable suction-based attachment system can be found in remora fish. Remora fish use suction pads—highly modified dorsal fin spines—to attach to sharks, whales, and manta rays (Beckert et al., 2015; Fulcher and Motta, 2006). Recent studies on the functional morphology of remora suction pads have greatly expanded our understanding on the mechanisms underlying their impressive performance (Beckert et al., 2015; Wang et al., 2017; Gamel et al., 2019). A remora suction pad comprises a soft fleshy outer rim and rows of lamellae topped with spinules. The pitch of the lamellae is muscle-controlled to facilitate spinule contact with rough surfaces, such as shark skin. When engaged, the tips of these stiff spinules interlock with surface asperities and increase friction, thereby increasing shear resistance. Moreover, the angled posterior-facing lamellae and spinules promote passive engagement when subjected to shear forces from a swimming host (Beckert et al., 2015; Fulcher and Motta, 2006). The similarities between the remora suction pad and the blepharicerid suction organ may be based on overlapping functional requirements, as both animals have to cope with high shear forces (fast-swimming hosts for the remora and torrential rivers for blepharicerid larvae).

Effect of the biofilm layer on suction attachments to natural rock surfaces

Since blepharicerid larvae attach to rocks underwater and feed on epilithic algae, their suction organs will in most cases contact the biofilm, yet the details of this interaction are unknown (Frutiger and Buergisser, 2002). As hypothesised previously, it is possible that the stiff microtrichia penetrate the biofilm layer (Rietschel, 1961; Nachtigall, 1974). This may allow the microtrichia to directly interlock with asperities on the rock surface or to generate additional friction from embedding numerous microtrichia into the biofilm. Mayfly larvae, which also inhabit fast-flowing watercourses, have friction-enhancing hairs that benefit from interacting with the biofilm (Ditsche et al., 2014a). It was found that a higher proportion of mayfly larvae can withstand fast flow rates on smooth hard (epoxy) substrates when the biofilm is present. Moreover, the setae and spine-like acanthae on the ventral surfaces of mayfly larvae can generate friction forces on clean rough substrates (Ditsche-Kuru et al., 2010). Additional experiments with blepharicerid larvae are underway to investigate the interaction between stiff microtrichia and soft substrates.

To conclude, we have shown that blepharicerid larvae use their suction organs to generate extreme attachment to diverse surfaces. The suction organ morphology is conserved between Hapalothrix and Liponeura larvae, and consists of a suction disc that contacts the substrate, dense arrays of microtrichia on the disc surface, muscles to control the piston, and the V-notch detachment system. We characterised the suction disc ultrastructure, which includes internal radial beam structures that could help to stabilise the suction disc when subjected to high stress,and fan-fibres that connect individual microtrichia to the radial beams. In terms of attachment performance, blepharicerid larvae withstand extreme shear forces equivalent to 320–1120 times their body weight on smooth substrates, depending on the species. H. lugubris performed the best overall, reaching shear forces equivalent to 1430 times the body weight, as well as an estimated shear stress of 117 kPa and normal stress of 24 kPa. Although their attachment decreased with increasing surface roughness, blepharicerid suction organs performed better than the smooth adhesive pads of stick insects on micro-rough surfaces. We confirmed that blepharicerid suction organs can mould to large surface asperities and that microtrichia come into close contact between the asperities. These microtrichia are stiff spine-like structures that are specialised for maintaining tip contact with the surface for interlocking with asperities. Our study provides new insights into the function of a highly adapted insect adhesive organ and expands our understanding of the function of biological suction organs.

Materials and methods

Sample collection and maintenance

Request a detailed protocol

L. cinerascens (Loew, 1845) larvae were collected from fast-flowing alpine rivers near Meiringen, Switzerland (GPS location 46° 44' 05.6" N, 8° 06' 55.4" E, in May 2018), and close to Grinzens, Tirol, Austria (47° 12' 41.4" N, 11° 15' 28.1" E, in September 2018). At the latter site, L. cordata (Vimmer, 1916) and H. lugubris (Loew, 1876) were also collected. For all the species, we collected third and fourth instar larvae that were large enough to be handled for experiments. Wearing fishing waders and diving gloves, we removed rocks from the most turbulent areas of the river and brought them to the riverbank for specimen collection. Although it was previously noted that the larvae can attach so firmly that they are torn upon detachment (Komárek, 1914), we found that a gentle nudge using soft-touch tweezers can elicit an evasive response from them, whereupon they could be easily picked up using tweezers and placed in specially prepared 50 ml Falcon tubes. All larvae were kept in these tubes in an ice box during collection and transport. Rocks were returned to their approximate locations after collection.

For long-term maintenance of the larvae, an aquarium tank was set up with water and small rocks from the collection site. A filter unit with two outlets for a small water cascade was used to filter the water and to simulate the natural environment. Multiple air pumps were also placed close to the aquarium walls to provide ample oxygenation and additional regions with turbulent flow. To promote algal growth, an over-tank light-emitting diode (LED) light was set to a 12-hr day-night cycle. The aquarium was kept in a 4°C climate room (mean temperature of 3.2°C ± 0.9°C; mean ± SD) to replicate alpine stream temperatures.

Scanning electron microscopy of Hapalothrix lugubris suction organs

Request a detailed protocol

SEM was used to image fourth instar H. lugubris larvae as described previously (Kang et al., 2019). In brief, samples fixed in 70% ethanol (v/v) were flash-frozen in liquid ethane cooled with liquid nitrogen and freeze-fractured immediately afterwards with a double-edged razor blade on a cooled aluminium block to obtain longitudinal views. Samples were freeze-dried overnight, then carefully mounted on SEM aluminium stubs using carbon tape and silver paint. They were then sputter-coated with 15 nm of iridium and imaged using a field-emission SEM (FEI Verios 460).

X-ray microtomography of blepharicerid suction organs

Request a detailed protocol

One H. lugubris fourth instar larva was fixed in 2% paraformaldehyde and 2% glutaraldehyde (v/v) in 0.05 M sodium cacodylate buffer (pH 7.4) for 7 days at 4°C. The larva was then dissected into six pieces—each containing one suction organ—and fixed for an additional day. The samples were then rinsed multiple times in 0.05 M sodium cacodylate buffer followed by deionised water before dehydration through a graded ethanol series: 50%, 75%, 95%, 100% (v/v), and 100% dry ethanol. The dehydrated samples were critical-point dried using four flushes of liquid CO2 in a Quorum E3100.

One critical-point-dried suction organ was used for imaging via micro-CT. The sample was mounted on a standard dressmaker’s pin using ultraviolet (UV)-curable glue, then imaged using a lab-based Zeiss Xradia Versa 520 (Carl Zeiss XRM, Pleasanton, CA, USA) x-ray microscope. The sample was scanned at 0.325 µm/pixel with an accelerating X-ray tube voltage of 50 kV and a tube current of 90 µA. A total of 2401 projections collected at 20 s exposure intervals were used to perform reconstruction using a Zeiss commercial software package (XMReconstructor, Carl Zeiss), utilising a cone-beam reconstruction algorithm based on filtered back-projection. Subsequent 3D volume rendering and segmentations were carried out using Dragonfly v4.0 (Object Research Systems Inc, Montreal, Canada) and Drishti v2.6.5 and v2.7 (Limaye and Stock, 2012).

Measuring attachment performance of blepharicerid larvae using a centrifuge

Request a detailed protocol

Insect attachment forces were measured using a custom centrifuge set up described previously (Federle et al., 2000). The centrifuge operated on the following principle: a platform with the test substrates and the insect was driven by a brushless motor, and a light barrier sensor was triggered per rotation. This signal was used to synchronise image acquisition from a USB camera (DMK 23UP1300; The Imaging Source Europe GmbH, Bremen, Germany), and image frames and their corresponding times were recorded using the StreamPix4 software (NorPix Inc, Montreal, Canada). For safety reasons, the maximum centrifugation speed was limited to approximately 75 rotations per second (rps). Some of the blepharicerid larvae could not be detached even at the maximum speed (n = 14 out of 136 measurements); in such cases, we used the maximum acceleration of a successfully detached individual from the given species.

Effect of surface roughness on the peak shear force of blepharicerid larvae was measured on the following substrates: smooth (clean polyester films), micro-rough (polishing films with a nominal asperity size of 0.05 µm; Ultra Tec, CA, US), and coarse-rough (30 µm polishing films; Ultra Tec). The same substrate types were used to measure the normal force (substrates mounted vertically in the centrifuge), but a polished polymethyl methacrylate (PMMA) surface was used as the smooth substrate. Surface characteristics (average roughness (mean height deviation) Ra, root-mean-squared roughness Rq, and maximum peak-to-valley height (PV)) of the micro-rough substrates were obtained using white-light interferometry with a scan area of 0.14 × 0.10 mm (Zygo NewView 200; Zygo Corporation, CT, USA). Micro-rough substrates were sputter-coated with 5 nm of iridium prior to scanning to improve the surface reflectivity. As the coarse-rough substrate could not be adequately imaged via white-light interferometry, we used a Z-stack image focal-depth analysis technique as described elsewhere (Sarmiento-Ponce et al., 2018) with a scan area of 0.44 × 0.58 mm. For both surfaces, three regions were selected at random and imaged. Interferometry images were analysed using MetroPro software (Zygo), and a custom MATLAB script was used to reconstruct the surface profile from the Z-stack images (The MathWorks Inc, MA, United States).

Since L. cinerascens and L. cordata were difficult to maintain in laboratory conditions, these two species were tested only on smooth horizontal surfaces (n = 43 and n = 10, respectively). The full range of tests on smooth, micro-rough, and coarse-rough surfaces was conducted for H. lugubris (n = 9–15 for shear tests; n = 10 for all tests in normal direction). Prior to the experiments, individuals were selected from the laboratory aquarium and placed inside specially prepared 50 ml Falcon tubes. This tube was kept on ice for the duration of the experiment. For each run, a larva was carefully removed from the tube and placed on the test surface. A droplet of water (taken from the aquarium) was used to wash excess debris from the insect, and lab tissue paper was used to wick away excess water without removing all moisture from the larva; the contact zone of the suction discs was still completely wetted under these conditions, as confirmed by IRM observations. These steps were necessary to prime the larvae for the centrifugation trials as they often displayed defensive behaviour while being handled. The larvae adhered and remained still once primed, and between two and four repetitions were performed for each larva. All centrifuge trials were conducted within 7 days of collection.

After the trials, all the larvae were blot-dried on filter paper and weighed using an analytical balance (1712 MP8; Sartorius GmbH, Göttingen, Germany). Statistical analyses were conducted on log10-transformed values using R v3.6.2 run in RStudio v 1.2.5033 (R Development Core Team, 2019; Team RS, 2019).

Calculating peak stress values on smooth horizontal plastic surfaces

Request a detailed protocol

Normal (adhesive) stress and shear stress (defined as the peak attachment force divided by the contact area) were calculated using suction disc areas measured for L. cordata and H. lugubris larvae. Larvae were placed on microscope slides so that the suction organs fully contacted the glass and imaged with a stereomicroscope. Every tested L. cordata and H. lugubris specimen was imaged. A representative organ was selected from each L. cordata and H. lugubris larva, and the contact area calculated by fitting a circle inclusive of the outer fringe layer using FIJI (Schindelin et al., 2012) (https://imagej.net/Fiji). The peak attachment force was then divided by this contact area to determine the peak stress for each larva.

Measuring peak shear and normal attachment forces of stick insects

Request a detailed protocol

C. morosus (Sinéty, 1901) stick insects were used as a model for terrestrial insect adhesion, and their attachment on surfaces with varying roughness was measured to compare against blepharicerid larval attachment. Second instar nymphs with undamaged legs and tarsi were selected for centrifuge experiments using smooth, micro-rough, and coarse-rough surfaces (n = 10 per surface). No normal forces could be measured on the micro-rough surface as stick insects failed to hold their body weight during preliminary tests. Before each trial, we checked that the specimen was in contact with the surface using all six legs and that the surface was uncontaminated. Stick insects were oriented with the head facing out, and each individual was tested twice and weighed afterwards. The higher attachment force per individual was used as the peak attachment force.

In vivo observation of suction organs attaching to smooth and micro-patterned substrates

Request a detailed protocol

In order to examine blepharicerid larvae locomoting for extended periods of time, a custom flow chamber was built to imitate the fast-flow conditions of their natural environments (Figure 8). Two aluminium plates (approximately 60 × 100 mm in height × width), each with a rectangular window, were used to sandwich an inner chamber made out of polydimethylsiloxane (PDMS; Sylgard 184, Dow Corning, MI, USA). This inner chamber had a lemon-shaped chamber to serve as the observation arena, and an inlet and an outlet for water circulation. Two microscope coverslips (0.16–0.19 mm thickness; Agar Scientific, Stansted, UK) were used to encase the inner chamber. Two to five larvae were placed on the bottom coverslip, and once the top coverslip was placed over the arena, four clamps were used to squeeze the aluminium plates and coverslips against the PDMS. The soft PDMS moulded closely to the plates and created a water-tight seal. Aquarium water (kept cool in an ice bath) was pumped via a micro-pump (M200S-V; TCS Micropumps Ltd, UK), and the input voltage was controlled by a microprocessor. The flow rate was controlled by setting an appropriate pump-operating voltage. With this flow chamber, we recorded H. lugubris larvae locomotion and the attachment/detachment of suction organs on smooth glass surfaces via IRM, which has been previously used to investigate the contact between animal adhesive organs and the substrate (Federle et al., 2006; Federle et al., 2002). Videos were recorded using a USB camera (DMK 23UP1300) and the IC Capture software (v2.4.642.2631; The Imaging Source GmbH) at 30 frames per second (FPS).

Figure 8
Download asset Open asset
Schematic of the flow chamber used to observe blepharicerid larvae locomoting in fast-flow conditions.

Two to five larvae at a time were added to the observation arena, covered with the top and bottom coverslips and plates, and imaged using interference reflection microscopy. A water pump continuously circulated cooled water at flow rates ranging between approximately 6 and 15 ml/s. Coverslip thickness: 0.16–0.19 mm. Plate dimensions: 100 × 60 mm in length × width. Observation arena dimensions: approximately 16 × 8 mm in width × height. PDMS: polydimethylsiloxane.

To observe how suction organs respond to surface roughness, we used transparent micro-structured surfaces with well-defined micro-ridges and grooves fabricated by photolithography and nanoimprinting (Zhou et al., 2014). In brief, a master surface was first produced using photolithography, and a PDMS mould of this master was used to cast the final surface out of epoxy. Three micro-ridge geometries were used in our experiments: (1) 3 × 3 × 2 µm (ridge width × groove width × ridge height); (2) 3 × 3 × 4 µm; and (3) 10 × 10 × 2 µm. As the ridge height is only approximately controlled through the spin-coating of photoresist when producing the master, we measured it from the epoxy replicas using white-light interferometry as mentioned above (see Table 1). Four to five regions from each uncoated substrate were imaged and only regions without artefacts were used in the final calculation. For simplicity, when referring to the substrates, the depths of the grooves were reported to one significant figure (ie, 3 × 3 × 2 µm, 3 × 3 × 4 µm, and 10 × 10 × 2 µm, for widths of ridges, grooves, and ridge height). Note that as these surfaces could not be used in combination with the flow chamber, a H. lugubris larva was placed on the substrate, wetted with a droplet of aquarium water, gently motivated with soft-touch forceps, and recorded as they moved around on the substrate. One H. lugubris specimen was used for both the 3 × 3 × 2 µm and 10 × 10 × 2 µm substrates, and a different larva was used for the 3 × 3 × 4 µm surface.

Estimating the elastic modulus of the microtrichia cuticle

Request a detailed protocol

Based on our observation that the microtrichia never showed any side contact, even when the suction discs were in contact with micro-structured substrates, we estimated the minimum elastic modulus of the microtrichia cuticle. We estimated the maximum force F on one microtrichium, perpendicular to the surface, as 56 nN. This was obtained based on the following assumptions: (1) the suction discs are loaded with a peak normal force of 11.6 mN, ie, 1.9 mN per sucker (from centrifuge measurements of H. lugubris); (2) equal loading of ca 34,000 spine-like microtrichia in tip contact (based on the area of the suction disc bearing spine-like microtrichia of ca 49,000 µm2, with the average microtrichia tip density of 0.7 per µm2). The microtrichia were assumed to be cylindrical, with a length L=6.7±0.5μm and diameter D=0.56±0.01μm (mean of means ± standard error of the mean; n = 2 H. lugubris), and the angle between the unloaded microtrichia and the surface was estimated as α=45±3° (mean ± SD measured from five microtrichia of H. lugubris). The local adhesion and friction force of the microtrichium were assumed to be negligible.

Following Goss and Chaouki, 2016, the elastic modulus below which a cylindrical beam loaded at an angle α (see ϕa definition) would exhibit side contact is

E=1Kp2-Fϕap22FL2I

where Kp2 and Fϕap2 are the complete and incomplete elliptic integrals of the first kind, respectively, p2=1/2 is the elliptic modulus, ϕa=sin-1sinα/2p is the elliptic amplitude, and I=D4π/64 is the second moment of area. See Figure 6d–ii for a schematic of a hypothetical scenario where microtrichia make side contacts on a smooth surface.

Data availability

Micro-computed tomography data is available in Dryad repository https://doi.org/10.5061/dryad.9zw3r22c2. Source data files have been provided for Figures 3, 4, and 5.

The following data sets were generated
    1. Kang V
    (2020) Dryad Digital Repository
    Supplementary dataset for Extreme suction attachment performance from specialised insects living in mountain streams (Diptera: Blephariceridae).
    https://doi.org/10.5061/dryad.9zw3r22c2

References

    1. Dai Z
    2. Gorb SN
    3. Schwarz U
    (2002)
    Friction of the beetle claw system
    The Journal of Experimental Biology 2002:2479–2488.
    1. Federle W
    2. Rohrseitz K
    3. Hölldobler B
    (2000)
    Attachment forces of ants measured with a centrifuge: better “wax-runners” have a poorer attachment to a smooth surface
    The Journal of Experimental Biology 203:505–512.
    1. Federle W
    2. Labonte D
    (2019) Dynamic biological adhesion: Mechanisms for controlling attachment during locomotion
    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 374:20190199.
    https://doi.org/10.1098/rstb.2019.0199
  31. Conference
    1. Komárek J
    (1914)
    Sitzungsberichte der Königlich Böhmischen Gesellschaft der Wissenschaften II Cl
    Die Morphologie und Physiologie der Haftscheiben der Blepharoceridenlarven. pp. 1–28.
    1. Komárek J
    2. Wimmer A
    (1922)
    The Larvae of the European Blepharoceridae (Diptera)
    Annales de Biologie Lacustre 22:63–77.
  34. Conference
    1. Limaye A
    2. Stock SR
    (2012) SPIE Optical Engineering + Applications
    Drishti: A volume exploration and presentation tool.
    https://doi.org/10.1117/12.935640
    1. Mannheims BJ
    (1935)
    Beiträge zur Biologie und Morphologie der Blepharoceriden (Dipt
    Zool Forschungen 1:115.
  37. Book
    1. Nachtigall W
    (1974)
    Biological Mechanisms of Attachment: The Comparative Morphology and Bioengineering of Organs for Linkage, Suction, and Adhesion
    Springer-Verlag.
  40. Software
    1. R Development Core Team
    (2019) R: A Language and Environment for Statistical Computing
    R Foundation for Statistical Computing, Vienna, Austria.
    http://www.r-project.org
  42. Book
    1. Roskov Y
    2. Ower G
    3. Nicolson D
    4. Bailly N
    5. Kirk PM
    6. Bourgoin T
    (2020)
    Species 2000 & ITIS Catalogue of Life, 2020-06-04 Beta
    Species 2000.
    1. Smith AM
    (1996)
    Cephalopod sucker design and the physical limits to negative pressure
    The Journal of Experimental Biology 199:949–958.
    1. Zwick P
    (2004)
    Insecta: Diptera, Blephariceridae
    Academy of Sciences Malaysia 10:736–749.

Decision letter

  1. David Lentink
    Reviewing Editor; University of Groningen, Netherlands
  2. George H Perry
    Senior Editor; Pennsylvania State University, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

Kang et al. eloquently describe the active suction organ that the larvae of aquatic insects of the dipteran family Blephariceridae use to adhere robustly to complex surfaces. While the morphology of the mechanism has been reported previously, its biomechanical adhesion function and performance across different substrates has been unknown. The authors present three advances. First, adhesion performance on rough, micro-rough, and smooth surfaces is quantified using an effective centrifugal setup. The performance tests show the larvae can resist shear forces up to 1100 times their body weight on smooth surfaces. Second, the suction function is visualized in vivo using interference reflection microscopy. This reveals that small hair-like microtrichia can enter gaps in the surface. Because the microtrichia are angled inward, the authors hypothesize that the microtrichia's angle and small size helps with interlocking and increasing suction-based attachment on rough surfaces. Finally, the adhesion performance of the Blephariceridae larvae is compared to other species, revealing that the weight-specific shear attachment on smooth surfaces is 3-10 times greater than found in stick insects. The finding that the larvae have such high attachment forces is impressive and the study offers new biological insights that may inspire engineers to invent new underwater suction mechanisms.

Decision letter after peer review:

Thank you for submitting your article "Extreme suction attachment performance from specialised insects living in mountain streams (Diptera: Blephariceridae)" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and George Perry as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission. Please comply with the comments and suggestions to your best ability, and respond in a point-by-point fashion. This is essential to enable the reviewing editor to fully evaluate the merit of your revision.

As the editors have judged that your manuscript is of interest, but as described below that additional work is required before it is published, we would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). First, because many researchers have temporarily lost access to the labs, we will give authors as much time as they need to submit revised manuscripts. We are also offering, if you choose, to post the manuscript to bioRxiv (if it is not already there) along with this decision letter and a formal designation that the manuscript is "in revision at eLife". Please let us know if you would like to pursue this option. (If your work is more suitable for medRxiv, you will need to post the preprint yourself, as the mechanisms for us to do so are still in development.)

Summary:

Kang et al. eloquently describe the active suction organ that the larvae of aquatic insects of the Dipterian family Blephariceridae use to adhere robustly to complex surfaces. While the morphology of the mechanism has been reported previously, it's biomechanical adhesion function and performance across different substrates is unknown. The authors present three advances. First, they quantify the adhesion performance on rough, micro-rough, and smooth surfaces using an effective centrifugal setup. The ultimate adhesion tests show the larvae can resist shear forces up to 1100 times their body weight on smooth surfaces. Second, they visualize the suction function in vivo using interference reflection microscopy. This reveals that small hair like microtrichia can enter gaps in the surface. Because the microtrichia are angled inward, the authors surmise that the microtrichia's angle and small size helps increase adhesion contact area on rough surfaces. Finally, they compare the adhesion performance of the Blephariceridae larvae to other species, showing it is 3-10 times greater than found in stick insects. The finding that the larvae have such high attachment forces is impressive and the study offers new biological insights that may inspire engineers to invent new underwater suction mechanisms.

Essential revisions:

Although the reviewers were generally appreciative of the well-written manuscript and the remarkable performance reported for the active suction mechanism, the consensus is that the mechanism itself is not described in sufficient detail for the reader to fully appreciate the advance. Hence the main critiques focus on helping the authors to further flesh out the mechanism and report it in more mechanistic detail like how other adhesion mechanism are described functionally across the biomechanical literature. Further the presentation of the figures does not meet graphic design clarity standards essential to inform eLife's broad readership. To provide guidance, we list the following essential revisions.

1. The introduction states that the suction organs have been observed, however, the manuscript does not communicate the observed mechanism as one would expect in the biomechanical adhesion literature. Instead it reports the measurements of the force and a suggestion that the microtrichia may be involved. We were hoping to find a quantitative report of the mechanism integrating the force data and microscopy images into biomechanical diagrams and to the extent possible, equations, that capture and communicate the mechanism as quantitatively as possible. Whereas we are not requesting further measurements, because the performance of the mechanism is well documented, we do ask a more in-depth biomechanical analysis that spells out the mechanism in a way it can be compared to the other classic mechanisms that the authors compare to. If this requires some additional measurements to inform the model, those efforts would be well worth it. In case the authors can use a mechanistic analysis lead, we recommend reviewing a couple of papers. E.g. Jeffries, Lindsie, and David Lentink. "Design Principles and Function of Mechanical Fasteners in Nature and Technology." Applied Mechanics Reviews 72.5 (2020). Or any other review or research paper that the authors find more useful.

2. Please clarify if the experiments are done in air or underwater. We consider underwater as most appropriate; at minimum the surface should be wetted. The authors mention that the Stefan adhesion forces underwater would be higher than in air, but it's not clear if that statement pertains to the experiment. Please provide a full clarification, and in case the experiments were performed in air we would prefer to see them performed in water. If this is not possible, the manuscript should be entirely transparent on this matter so the reader can evaluate the precise merit of this study and its limitations fully.

3. We found the images confusing at times. To resolve this we would like to see clear schematics (avatars) that ground the reader's perspective in all figures.

4. Considering eLife's broad multidisciplinary readership and the appeal of this study for bioinspired designers and engineers, Figure 1d,e has to provide better anatomical readability. Please assume a Biology and Engineering undergrad level for the first figure, ensuring all definitions and anatomical names can be fully comprehended without reference to other literature. Please provide clear connections to the different views and perspectives presented in the panels leveraging graphic design to the benefit of the interested reader not familiar with insect morphology.

5. Likewise, Figure 2 is also confusing. A schematic is in order to show the reader what they are looking at, how the images relate, and why they matter (significance) for understanding the main findings reported in this manuscript.

6. Figure 3 clearly shows that course-rough surfaces provide far less adhesion force. We wonder, are there any images similar to Figure 6 showing that the microtrichia cannot enter the gaps? To comprehend what causes the differences, we would like to see a report of the length scale of the microtrichia compared to that of the gap's dimensions, both for the rough and micro rough surfaces. To clarify this in a universal fashion, please consider reporting gap size non-dimensionally based on the relevant microtrichia length scale. More discussion of the relevant length scales would help bring the force measurements and the observations of the microtrichia together.

7. Figure 6 is an important figure, so it would help the reader to more easily grasp the viewing perspective using diagrams and avatars. I panel a, a schematic should clearly define the suction disc fringe and the perspective shown. What part is the suction disc and what is the length scale of this image compared to the suction disc? Also, it would be useful if the columns of the microstructure could all be aligned for clarity.

8. Currently, the authors provide an estimate of the shear stress. It would be helpful to also include the normal stress based on the normal force data on smooth surfaces for lugubris. It would be informative for the reader to know if it exceeds 1 atm. If so, that is a very interesting finding. Please report and discuss what you find in the revised manuscript.

9. Discussion: Please include a comparison of the magnitude of shear and normal stress that this suction mechanism creates with that of other organisms. Currently the comparison is done with force per body weight, which is biologically relevant. However, reporting stress provides an objective bio-mechanistic perspective on adhesion performance.

10. Discussion, Ln 300: The suggestion that the inward-facing microtrichia may function to prevent inward slipping of the suction cup is interesting. Please discuss the tradeoff between smooth and micro-rough surfaces: is it possible that on micro-rough surfaces the microtrichia are better able to resist slip, but on smooth surfaces, the seal is better? And if so, this would suggest the effect of a better seal is more important than preventing slip, since performance is better on smooth surfaces? In-vivo visualization during failure would be very informative (in future work).

11. Please discuss why there may be an intricate branching of the fan-fibres into the microtrichia. E.g. in the gecko, the branched tendons insert into the lamella, supporting the large tensile loads applied to the adhesive. However, here it is less clear if large tensile loads would be applied to the microtrichia. It seems logical that applying large normal loads to the suction cup should be done at its centre, resulting in decreased pressure if no slip occurs (as opposed to applying the normal force to the rim, which would not decrease pressure). So, this would not explain the intricate network of fan-fibres. However, for shear loads, it could make more sense: pulling in shear would engage the microtrichia on the far side of the cup, and the fan-fibres could help transmit this tension. It might be worth thinking this through and discussing the outcome in the paper to strengthen the mechanistic analysis.

12. We would be excited to learn if the authors have thoughts on the slight curvature of the microtrichia and how it may be involved in the adhesion mechanism. In case this is purely speculative, this could go into the last paragraph of the paper, alternatively it could go into the biomechanical model of the mechanism.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for submitting your article "Extreme suction attachment performance from specialised insects living in mountain streams (Diptera: Blephariceridae)" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and George Perry as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission. Please comply with the comments and suggestions to your best ability, and respond in a point-by-point fashion. This is essential to enable the reviewing editor to fully evaluate the merit of your revision. Upon evaluating your current revision, we noted these expectations were not fully met. To evaluate the revision, we had to request another review cycle, and this uncovered some remaining issues. We offer the authors this final chance to comply and transparently communicate the manuscript changes.

Essential revisions:

1) The videos and images are striking but difficult to interpret. A schematic or two of the entire suction disc organ would be particularly useful; with the various parts (piston, v-shaped notch) all labelled clearly. That will make it easier to find them in the SEM images, since the SEMS have several different views and many detail that obscure the main parts.

2) Please draw the microtrichia and supporting beams in the schematic to make the images in Figure 2 easier to interpret for eLife's broad readership.

3) The authors compare normal to shear stress, but the reader wasn't informed about normal stress yet. The authors should clarify earlier that shear and normal directions will be tested. The justification why the study focusses on shear stress should be presented earlier to help the reader get the main point. For example, this could be accomplished by including a view of the larvae on a rock in a stream showing the natural conditions under which they might be dislodged. Even a schematic would do. Presenting this early, e.g. "Figure 1a", would make all the difference for eLife's broad readership.

4) The authors model predicts that the microtrichia cuticle has a Young's modulus is 0.3 GPa. Please explain what the composition of the cuticle is and how the Young's modulus compares to the stiffness of similarly composed biomaterials.

5) The description of various microtrichia scenarios is confusing for the general reader. A diagram could help (or otherwise consider deleting this section). We were hoping to find a quantitative report of the mechanism integrating the force data and microscopy images into biomechanical diagrams and to the extent possible, equations, that capture and communicate the mechanism as quantitatively as possible. Ideally, the model would capture the basic results seen in the force data, for instance, explaining the change in force as surface roughness varied. In case this is outside the scope of the study, please discuss this open avenue of research in one of the closing sections of the manuscript. So, the reader is alerted that the mechanism isn't fully understood yet and which further research steps are needed to close this gap.

6) The authors should either show a clarifying schematic (preferred) or present a clear description of "beam side contact".

7) Figure 3 – the avatars are helpful; to fully clarify this figure, please draw a symbol to clearly indicate the axis of spin of the device as well.

8) The insets are very helpful. Many readers will, however, remain confused by the orientations of the images. Is the V-notch anterior or posterior? In 1b, it appears to be located on the left side of the discs, but in 1a, left appears to be anterior? Then in 2a, the inset shows the disc with what appears to be the V-notch located at the top of the inset, which is different than the V-C-D reference frame in the image? The fact that the slice is at 45 degrees, but then the reference frame seems to be parallel to the slice seems odd. What is most important for the reader is to know in 2a which direction is toward the center of the disc and which is toward the edge. It is not currently clear. Finally, in 2f, it now appears as though the disc has flipped orientations with respect to the inset? Is it possible to maintain the same orientation? Please resolve these issues holistically for the general eLife reader.

9) 6a-iii appears to be a different location than 6a-ii? If so, please add a schematic here as well to well-inform the general eLife reader. Please note 6-a-iii is missing a label.

10) Despite the additional explanation, the following two items that are unclear:

A -"To function effectively and to avoid buckling in this situation, interlocking structures…"

Buckling is usually defined as a sudden change in shape under an increasing load, often a beam loaded axially. Here there is a side load on the beam, so is this simply transverse bending rather than buckling? Or is the load on the side inducing buckling, please clarify what you would like to confer to the reader by thoughtfully considering the connotations.

B -"A second effect of the microtrichia curvature is that bending and thus side contact of the fine microtrichia tips is avoided, which would again reduce their ability to interlock with the substrate."

Is bending avoided via curvature? Or is it just that under a given amount of bending, having a pre-curve results in a shape that does not make side-contact? Please clarify so the general eLife reader can follow.

https://doi.org/10.7554/eLife.63250.sa1

Author response

Essential revisions:

Although the reviewers were generally appreciative of the well-written manuscript and the remarkable performance reported for the active suction mechanism, the consensus is that the mechanism itself is not described in sufficient detail for the reader to fully appreciate the advance. Hence the main critiques focus on helping the authors to further flesh out the mechanism and report it in more mechanistic detail like how other adhesion mechanism are described functionally across the biomechanical literature. Further the presentation of the figures does not meet graphic design clarity standards essential to inform eLife's broad readership. To provide guidance, we list the following essential revisions.

1. The introduction states that the suction organs have been observed, however, the manuscript does not communicate the observed mechanism as one would expect in the biomechanical adhesion literature. Instead it reports the measurements of the force and a suggestion that the microtrichia may be involved. We were hoping to find a quantitative report of the mechanism integrating the force data and microscopy images into biomechanical diagrams and to the extent possible, equations, that capture and communicate the mechanism as quantitatively as possible. Whereas we are not requesting further measurements, because the performance of the mechanism is well documented, we do ask a more in-depth biomechanical analysis that spells out the mechanism in a way it can be compared to the other classic mechanisms that the authors compare to. If this requires some additional measurements to inform the model, those efforts would be well worth it. In case the authors can use a mechanistic analysis lead, we recommend reviewing a couple of papers. E.g. Jeffries, Lindsie, and David Lentink. "Design Principles and Function of Mechanical Fasteners in Nature and Technology." Applied Mechanics Reviews 72.5 (2020). Or any other review or research paper that the authors find more useful.

We thank the reviewers for highlighting this gap in our manuscript. After reading the suggested paper and reassessing our manuscript, we decided to include a new theoretical model and a schematic to further support our findings on the function of microtrichia during suction attachments (Figure 7, with accompanying text in Discussion lines 287-299, and Materials and methods lines 533 – 552). In order for the microtrichia to effectively interlock with surface asperities, they need to have sufficient stiffness to resist bending and buckling. Using our model, we estimated the elastic modulus of the microtrichia cuticle necessary to prevent side contact to be 0.3 – 0.4 GPa or higher, which is a range typical of stiff sclerotised cuticle, –which can have elastic moduli as high as 20 GPa. This result, along with our in vivo observations of the microtrichia always making tip contact on smooth and microstructured surfaces and microscopy images revealing their solid (filled-in) cuticular ultrastructure, strongly indicates that the spine-like microtrichia are stiff structures that can improve suction performance by interlocking their tips with surface asperities and increasing friction near the disc rim.

2. Please clarify if the experiments are done in air or underwater. We consider underwater as most appropriate; at minimum the surface should be wetted. The authors mention that the Stefan adhesion forces underwater would be higher than in air, but it's not clear if that statement pertains to the experiment. Please provide a full clarification, and in case the experiments were performed in air we would prefer to see them performed in water. If this is not possible, the manuscript should be entirely transparent on this matter so the reader can evaluate the precise merit of this study and its limitations fully.

Thank you for the comment. We have added this sentence to Line 116 to clarify the experimental condition:

“Each specimen was wetted with a droplet of water prior to centrifugation (see Materials and methods for details). Interference reflection microscopy (IRM, see below) observations showed that the contact of the suction organs under these conditions was completely wet, and no air bubbles were present in the contact zone.”

Please note that, since the centrifuge force measurement depends on the insect's (unsubmerged) weight and requires high quality video recording to determine the precise time of detachment, it is not possible to do these experiments entirely underwater. Instead, we wetted the blepharicerid larvae immediately prior to centrifugation, as described in the Materials and methods section (lines 465 468). The contact of the suction organs under these conditions was completely wet (IRM observations showed that no air bubbles were present in the contact zone) and the results are therefore comparable to the performance of the suction organs under natural conditions. Air bubbles in the contact zone would also be inconsistent with the lowest suction pressures we observed, as such air bubbles would expand so much in volume that the suction discs would likely detach. The wetting of the body surface did not influence the measurements via the body weight, as we could see small water droplets being removed from the larva during the centrifugation, and the insects were weighed after the experiment.

3. We found the images confusing at times. To resolve this we would like to see clear schematics (avatars) that ground the reader's perspective in all figures.

Thank you for raising this issue. We have revised Figures 1, 2, and 6 accordingly, and have included a new figure (Figure 7) that portrays a mechanical model of how the microtrichia system could interlock and avoid buckling against rough substrates (outlined in detail in the response to #1).

4. Considering eLife's broad multidisciplinary readership and the appeal of this study for bioinspired designers and engineers, Figure 1d,e has to provide better anatomical readability. Please assume a Biology and Engineering undergrad level for the first figure, ensuring all definitions and anatomical names can be fully comprehended without reference to other literature. Please provide clear connections to the different views and perspectives presented in the panels leveraging graphic design to the benefit of the interested reader not familiar with insect morphology.

We acknowledge that the previous layout of Figure 1 may have been confusing to non-specialists. We have revised Figure 1 by incorporating additional schematics and labels. In addition, please note that we included a Rich Media File (Video 1) during the initial submission that showcases the 3D rendering based on micro-CT data. Since Figure 1d and 1e are based on this dataset, we hope that this Rich Media File will further clarify the morphology to the reader.

5. Likewise, Figure 2 is also confusing. A schematic is in order to show the reader what they are looking at, how the images relate, and why they matter (significance) for understanding the main findings reported in this manuscript.

We have added a schematic of the suction disc to better orient the reader (see Figure 2). This is in addition to the changes made in Figure 1, and we believe this combination improves clarity for the reader.

6. Figure 3 clearly shows that course-rough surfaces provide far less adhesion force. We wonder, are there any images similar to Figure 6 showing that the microtrichia cannot enter the gaps? To comprehend what causes the differences, we would like to see a report of the length scale of the microtrichia compared to that of the gap's dimensions, both for the rough and micro rough surfaces. To clarify this in a universal fashion, please consider reporting gap size non-dimensionally based on the relevant microtrichia length scale. More discussion of the relevant length scales would help bring the force measurements and the observations of the microtrichia together.

Unfortunately, it is not possible to visualise the contact of the suction organs on the micro and coarse rough substrates since they are not transparent (the substrates have a polyester backing layer). To get around this problem, we used the epoxy-based microstructured substrates. The length-scales of the microtrichia (mean length 6.7 µm) and of the rough test surfaces (peak-to-valley heights: micro-rough 4.56 µm, coarse-rough 78.82 µm) are given in Table 1, Figure 7, and in the text. The ability of the suction organ to reach into the gaps between the asperities depends not only on the length of the microtrichia but also on the flexibility of the suction disc as a whole, as it can deform around the larger structures. For clarification, we have added an explanation to the Discussion (lines 226 – 229): "It is likely that both blepharicerid suction organs and stick insect adhesive pads are unable to cope with coarse surface roughness. The adhesive pads of both insects may be unable to fully mould to the large asperities, and the length of the microtrichia may be insufficient to reach the lower regions of the surface profile (Figure 6).”

In addition, we have revised Figure 6 to include schematics that clearly illustrate the length-scales of the microtrichia in comparison to the microstructured surfaces.

7. Figure 6 is an important figure, so it would help the reader to more easily grasp the viewing perspective using diagrams and avatars. I panel a, a schematic should clearly define the suction disc fringe and the perspective shown. What part is the suction disc and what is the length scale of this image compared to the suction disc? Also, it would be useful if the columns of the microstructure could all be aligned for clarity.

Thank you for the suggestions. We have revised the figure so that the sub-panels are aligned, and we have also added a schematic to highlight the approximate location and size of the images with respect to the overall suction disc (Figure 6a-ii). Moreover, we added scaled schematics to convey the length-scales of the microtrichia with respect to the microstructured surfaces.

8. Currently, the authors provide an estimate of the shear stress. It would be helpful to also include the normal stress based on the normal force data on smooth surfaces for lugubris. It would be informative for the reader to know if it exceeds 1 atm. If so, that is a very interesting finding. Please report and discuss what you find in the revised manuscript.

We have revised Table 2 (line 561) to include both shear and normal stress estimates for L. cordata and H. lugubris. The two highest normal stress values were obtained from two different L. cordata larvae, at 73 and 77 kPa. These are conservative estimates, however, as (1) we were unable to detach them using the centrifuge method, so the maximum recorded forces prior to termination were used; (2) we assumed the contact area to include the outer fringe layer; (3) all six organs were assumed to remain in contact prior to detachment. If we use a contact area without the fringe layer, then the normal stresses would be 108 and 114 kPa. In addition, larvae often attach with fewer than six organs immediately prior to detachment (also described Frutiger A. 2002. The function of the suckers of larval net-winged midges (Diptera: Blephariceridae). Freshwater Biology 47:293–302), which quickly increases the stress values (e.g., if three organs were in contact, then stress values would be 216 and 228 kPa). Although a pressure difference of 101 kPa (1 atm at standard sea level and temperature) is considered the upper threshold for suction attachments in air, this is not the case underwater, where the high cohesive strength of water allows for pressure differences to exceed 1 atm. For example, octopus and squid suckers can withstand normal stresses greater than 101 kPa, reaching up to 271 kPa for the octopus and 830 kPa for squids (Smith AM. 1996. Cephalopod sucker design and the physical limits to negative pressure. J Exp Biol 199:949–58). Thus, blepharicerid suction organs, like those of the cephalopods, are able to withstand strong normal stresses that exceed 101 kPa.

We have incorporated elements of this text in the Discussion (lines 250 – 260) to address points #8 and 9.

9. Discussion: Please include a comparison of the magnitude of shear and normal stress that this suction mechanism creates with that of other organisms. Currently the comparison is done with force per body weight, which is biologically relevant. However, reporting stress provides an objective bio-mechanistic perspective on adhesion performance.

Please refer to our response to point #8 and the revised Discussion (lines 250 – 260).

10. Discussion, Ln 300: The suggestion that the inward-facing microtrichia may function to prevent inward slipping of the suction cup is interesting. Please discuss the tradeoff between smooth and micro-rough surfaces: is it possible that on micro-rough surfaces the microtrichia are better able to resist slip, but on smooth surfaces, the seal is better? And if so, this would suggest the effect of a better seal is more important than preventing slip, since performance is better on smooth surfaces? In-vivo visualization during failure would be very informative (in future work).

We believe both an effective seal and an increase in friction help with suction performance. For example, studies that tested bio-inspired suction cup designs on rough substrates (Wang Y et al. 2017. A biorobotic adhesive disc for underwater hitchhiking inspired by the remora suckerfish. Sci Robot 2:eaan8072; Ditsche P, Summers A. 2019. Learning from Northern clingfish (Gobiesox maeandricus): Bioinspired suction cups attach to rough surfaces. Philos Trans R Soc B Biol Sci 374:1784) found that the combination of a soft sealing margin and friction-enhancing microstructures leads to peak attachment performance on rough substrates. We agree, however, that without an efficient seal, the loss of suction performance will likely outweigh the gain from having friction-enhancing structures. We have incorporated elements of this text into the Discussion (lines 310 – 314).

11. Please discuss why there may be an intricate branching of the fan-fibres into the microtrichia. E.g. in the gecko, the branched tendons insert into the lamella, supporting the large tensile loads applied to the adhesive. However, here it is less clear if large tensile loads would be applied to the microtrichia. It seems logical that applying large normal loads to the suction cup should be done at its centre, resulting in decreased pressure if no slip occurs (as opposed to applying the normal force to the rim, which would not decrease pressure). So, this would not explain the intricate network of fan-fibres. However, for shear loads, it could make more sense: pulling in shear would engage the microtrichia on the far side of the cup, and the fan-fibres could help transmit this tension. It might be worth thinking this through and discussing the outcome in the paper to strengthen the mechanistic analysis.

Thank you for suggesting the interesting idea that the fan-fibres could serve to transmit forces during shear loads on the suction disc, which would pull interlocked microtrichia radially outward (away from the centre of the suction disc). As the direction of the fan-fibres in the default (unstrained) condition is perpendicular to the surface, however, these fibres could only transmit forces parallel to the surface after a substantial outward displacement of the interlocked microtrichia. Given that the radial extension of the fan-fibre zone is relatively small (ca 20 µm, see Figure 2a), and that the fan-fibres are up to 20 µm long, we think that shear force transmission via the fan-fibres is unlikely.

One possible function of the branched fan-fibres is to coordinate microtrichia movement. Tension on the fan-fibres could be achieved via a raised hydraulic pressure within the fan-fibre space. A possible hint that microtrichia movement is activated via the fan-fibres is the observation that the microtrichia often flickered in our IRM video recordings. It is also possible, however, that this movement is caused by the flow of water in and out of the suction organ.

Another possibility is that the presence of the fan-fibres is a consequence of the development of the microtrichia, which involves the formation of long cytoplasmic extensions of epidermal cells (Rietschel P. 1961. Bau, Funktion und Entwicklung der Haftorgane der Blepharoceridae. Z Morph Ökol Tiere 50:239–265). Since little is known about the detailed development of the microtrichia or the suction organ in general, additional work is needed to explore this hypothesis.

As the above arguments are highly speculatory, we would like to refrain from including them in the manuscript.

12. We would be excited to learn if the authors have thoughts on the slight curvature of the microtrichia and how it may be involved in the adhesion mechanism. In case this is purely speculative, this could go into the last paragraph of the paper, alternatively it could go into the biomechanical model of the mechanism.

Thank you for your suggestion. We have added the following paragraph to the Discussion (lines 301 – 308):

“The densely packed microtrichia are both slightly curved and tapered, and their base is much thicker than the tip. […] A second effect of the microtrichia curvature is that bending and thus side contact of the fine microtrichia tips is avoided, which would again reduce their ability to interlock with the substrate.”

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Essential revisions:

1) The videos and images are striking but difficult to interpret. A schematic or two of the entire suction disc organ would be particularly useful; with the various parts (piston, v-shaped notch) all labelled clearly. That will make it easier to find them in the SEM images, since the SEMS have several different views and many detail that obscure the main parts.

Thank you for your feedback. We have made extensive revisions to both Figure 1 and 2 to make them clearer to the general audience. In particular, we have included two additional sub-panels to Figure 1 (as well as a new video, Video 1).

In Figure 2, we have included a scanning electron micrograph giving an overview of the suction disc that is accompanied by a schematic of the suction disc, both of which are clearly labelled. We have also reoriented sub-panels so that the suction discs all face the same direction (see also our response to Comment #8). We hope these revisions will improve the overall reading experience.

2) Please draw the microtrichia and supporting beams in the schematic to make the images in Figure 2 easier to interpret for eLife's broad readership.

This has been addressed in the revised Figure 2a-i and ii.

3) The authors compare normal to shear stress, but the reader wasn't informed about normal stress yet. The authors should clarify earlier that shear and normal directions will be tested. The justification why the study focusses on shear stress should be presented earlier to help the reader get the main point. For example, this could be accomplished by including a view of the larvae on a rock in a stream showing the natural conditions under which they might be dislodged. Even a schematic would do. Presenting this early, e.g. "Figure 1a", would make all the difference for eLife's broad readership.

We have included an image (Figure 1a) and a video (Video 1) of the larvae in their natural habitat that demonstrate the need for powerful attachments to resist high shear forces. We agree that these additions provide more context to the readers and thank the reviewer for the suggestions.

In the previous version we used multiple terms to refer to adhesion force. We have now replaced "adhesion" or "adhesive force" with "normal force" where applicable to improve consistency and readability. Please note that we refer to both shear and normal forces before line 254: “Peak shear and

normal (adhesive) forces per body weight were measured on horizontal and vertical substrates, respectively (Figure 3a and b)” (see line 120 in the re-submitted version).

4) The authors model predicts that the microtrichia cuticle has a Young's modulus is 0.3 GPa. Please explain what the composition of the cuticle is and how the Young's modulus compares to the stiffness of similarly composed biomaterials.

We have included additional information on insect cuticle and comparable biomaterials in the revised paragraph (line 288 – 301).

5) The description of various microtrichia scenarios is confusing for the general reader. A diagram could help (or otherwise consider deleting this section). We were hoping to find a quantitative report of the mechanism integrating the force data and microscopy images into biomechanical diagrams and to the extent possible, equations, that capture and communicate the mechanism as quantitatively as possible. Ideally, the model would capture the basic results seen in the force data, for instance, explaining the change in force as surface roughness varied. In case this is outside the scope of the study, please discuss this open avenue of research in one of the closing sections of the manuscript. So, the reader is alerted that the mechanism isn't fully understood yet and which further research steps are needed to close this gap.

We added this description of the possible functions of the microtrichia shape (curvature and taper) in response to a comment from the first round of reviews. We believe this section can be removed as it is based on speculations.

In terms of a biomechanical model that can explain the suction organ attachment performance on various substrates: due to the highly complex morphology of the blepharicerid suction organ and the relative novelty of the study system, it is currently not possible to formulate a quantitative model to capture their performance on rough substrates. We acknowledge this limitation and have added the following sentence (line 307): “The detailed biomechanics of how microtrichia-covered blepharicerid suction organs produce a tight seal on rough surfaces and interlock with small substrate asperities is beyond the scope of this study and remains to be explored in future work.”

6) The authors should either show a clarifying schematic (preferred) or present a clear description of "beam side contact".

Thank you for the suggestion. We have added Figure 6d-i and d-ii to illustrate side contact, and have added the following sentence to line 561: “See Figure 6d-ii for a schematic of a hypothetical scenario where microtrichia make side contact on a smooth surface.”

7) Figure 3 – the avatars are helpful; to fully clarify this figure, please draw a symbol to clearly indicate the axis of spin of the device as well.

We have addressed this in the revised Figure 3.

8) The insets are very helpful. Many readers will, however, remain confused by the orientations of the images. Is the V-notch anterior or posterior? In 1b, it appears to be located on the left side of the discs, but in 1a, left appears to be anterior? Then in 2a, the inset shows the disc with what appears to be the V-notch located at the top of the inset, which is different than the V-C-D reference frame in the image? The fact that the slice is at 45 degrees, but then the reference frame seems to be parallel to the slice seems odd. What is most important for the reader is to know in 2a which direction is toward the center of the disc and which is toward the edge. It is not currently clear. Finally, in 2f, it now appears as though the disc has flipped orientations with respect to the inset? Is it possible to maintain the same orientation? Please resolve these issues holistically for the general eLife reader.

We acknowledge that the orientations in the previous version of Figure 1 and 2 could have been more consistent. We have addressed this issue with additional direction markers for Figure 1c and d, as well as re-orienting Figure 2 sub-panels so that both the images and the schematics consistently face the same direction (left being anterior).

9) 6a-iii appears to be a different location than 6a-ii? If so, please add a schematic here as well to well-inform the general eLife reader. Please note 6-a-iii is missing a label.

We have added a schematic to 6a-iii to illustrate the imaged region of the suction disc. We have also added a label.

10) Despite the additional explanation, the following two items that are unclear:

A -"To function effectively and to avoid buckling in this situation, interlocking structures…"

Buckling is usually defined as a sudden change in shape under an increasing load, often a beam loaded axially. Here there is a side load on the beam, so is this simply transverse bending rather than buckling? Or is the load on the side inducing buckling, please clarify what you would like to confer to the reader by thoughtfully considering the connotations.

We realise that this sentence (“To function effectively and to avoid buckling in this situation, interlocking structures…”) is confusing as we are not referring to the buckling of the microtrichia but to the buckling of the disc rim, which is a common mechanism of failure in underwater suction attachments (see line 264 in re-submitted version). We have revised the sentence to remove this ambiguity: “To interlock effectively, structures like the remora spinules need to be stiff and strong [38,39].”

B -"A second effect of the microtrichia curvature is that bending and thus side contact of the fine microtrichia tips is avoided, which would again reduce their ability to interlock with the substrate."

Is bending avoided via curvature? Or is it just that under a given amount of bending, having a pre-curve results in a shape that does not make side-contact? Please clarify so the general eLife reader can follow.

We have opted to remove this section (line 309-317 in the reviewed version) as it was causing some confusion for the reviewers (see Comment #5). Furthermore, since we currently lack experimental support for our ideas, this section is mainly speculation based on morphology, and removing it will not impact the findings nor the overall story.

https://doi.org/10.7554/eLife.63250.sa2

Article and author information

Author details

  1. Victor Kang

    Department of Zoology, University of Cambridge, Cambridge, United Kingdom
    Present address
    Department of Bioengineering, Imperial College London, London, United Kingdom
    Contribution
    Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review and editing
    For correspondence
    k.kang@imperial.ac.uk
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0959-1364

  2. Robin T White

    Carl Zeiss Research Microscopy Solutions, Pleasanton, United Kingdom
    Contribution
    Methodology, Writing – original draft, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-0030-2872

  3. Simon Chen

    Department of Zoology, University of Cambridge, Cambridge, United Kingdom
    Contribution
    Methodology, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6316-7567

  4. Walter Federle

    Department of Zoology, University of Cambridge, Cambridge, United Kingdom
    Contribution
    Conceptualization, Funding acquisition, Methodology, Supervision, Writing – review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6375-3005

Funding

EU Horizon 2020 research and innovation programme (No. 642861)

  • Victor Kang
  • Walter Federle

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We would like to thank P Ladurner for assisting in sample collection and providing laboratory space near to the field site and M Sutcliffe for helping with the surface profilometry measurements. We are also grateful to KH Muller and JN Skepper at the Cambridge Advanced Imaging Centre for their help in preparing and imaging SEM samples.

Senior Editor

  1. George H Perry, Pennsylvania State University, United States

Reviewing Editor

  1. David Lentink, University of Groningen, Netherlands

Version history

  1. Received: September 18, 2020
  2. Preprint posted: October 2, 2020 (view preprint)
  3. Accepted: September 24, 2021
  4. Version of Record published: November 3, 2021 (version 1)

Copyright

© 2021, Kang et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 826
    Page views
  • 92
    Downloads
  • 7
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Victor Kang
  2. Robin T White
  3. Simon Chen
  4. Walter Federle
(2021)
Extreme suction attachment performance from specialised insects living in mountain streams (Diptera: Blephariceridae)
eLife 10:e63250.
https://doi.org/10.7554/eLife.63250

Categories and tags

Further reading

    1. Ecology
    2. Genetics and Genomics

    Adulis and the transshipment of baboons during classical antiquity

    Franziska Grathwol, Christian Roos ... Gisela H Kopp
    Research Advance

    Adulis, located on the Red Sea coast in present-day Eritrea, was a bustling trading centre between the first and seventh centuries CE. Several classical geographers--Agatharchides of Cnidus, Pliny the Elder, Strabo-noted the value of Adulis to Greco--Roman Egypt, particularly as an emporium for living animals, including baboons (Papio spp.). Though fragmentary, these accounts predict the Adulite origins of mummified baboons in Ptolemaic catacombs, while inviting questions on the geoprovenance of older (Late Period) baboons recovered from Gabbanat el-Qurud ('Valley of the Monkeys'), Egypt. Dated to ca. 800-540 BCE, these animals could extend the antiquity of Egyptian-Adulite trade by as much as five centuries. Previously, Dominy et al. (2020) used stable istope analysis to show that two New Kingdom specimens of P. hamadryas originate from the Horn of Africa. Here, we report the complete mitochondrial genomes from a mummified baboon from Gabbanat el-Qurud and 14 museum specimens with known provenance together with published georeferenced mitochondrial sequence data. Phylogenetic assignment connects the mummified baboon to modern populations of Papio hamadryas in Eritrea, Ethiopia, and eastern Sudan. This result, assuming geographical stability of phylogenetic clades, corroborates Greco-Roman historiographies by pointing toward present-day Eritrea, and by extension Adulis, as a source of baboons for Late Period Egyptians. It also establishes geographic continuity with baboons from the fabled Land of Punt (Dominy et al., 2020), giving weight to speculation that Punt and Adulis were essentially the same trading centres separated by a thousand years of history.

    1. Ecology

    Engineering multifunctional rhizosphere probiotics using consortia of Bacillus amyloliquefaciens transposon insertion mutants

    Jingxuan Li, Chunlan Yang ... Zhong Wei
    Research Article Updated

    While bacterial diversity is beneficial for the functioning of rhizosphere microbiomes, multi-species bioinoculants often fail to promote plant growth. One potential reason for this is that competition between different species of inoculated consortia members creates conflicts for their survival and functioning. To circumvent this, we used transposon insertion mutagenesis to increase the functional diversity within Bacillus amyloliquefaciens bacterial species and tested if we could improve plant growth promotion by assembling consortia of highly clonal but phenotypically dissimilar mutants. While most insertion mutations were harmful, some significantly improved B. amyloliquefaciens plant growth promotion traits relative to the wild-type strain. Eight phenotypically distinct mutants were selected to test if their functioning could be improved by applying them as multifunctional consortia. We found that B. amyloliquefaciens consortium richness correlated positively with plant root colonization and protection from Ralstonia solanacearum phytopathogenic bacterium. Crucially, 8-mutant consortium consisting of phenotypically dissimilar mutants performed better than randomly assembled 8-mutant consortia, suggesting that improvements were likely driven by consortia multifunctionality instead of consortia richness. Together, our results suggest that increasing intra-species phenotypic diversity could be an effective way to improve probiotic consortium functioning and plant growth promotion in agricultural systems.

    1. Computational and Systems Biology
    2. Ecology

    Gut Health: The value of connections

    Vanessa Rossetto Marcelino
    Insight

    High proportions of gut bacteria that produce their own food can be an indicator for poor gut health.