Convergent evolution of saccate body shapes in nematodes through distinct developmental mechanisms
The vast majority of nematode species have vermiform (worm-shaped) body plans throughout post-embryonic development. However, atypical body shapes have evolved multiple times. The plant-parasitic Tylenchomorpha nematode Heterodera glycines hatches as a vermiform infective juvenile. Following infection and the establishment of a feeding site, H. glycines grows disproportionately greater in width than length, developing into a saccate adult. Body size in Caenorhabditis elegans was previously shown to correlate with post-embryonic divisions of laterally positioned stem cell-like ‘seam’ cells and endoreduplication of seam cell epidermal daughters. To test if a similar mechanism produces the unusual body shape of saccate parasitic nematodes, we compared seam cell development and epidermal ploidy levels of H. glycines to C. elegans. To study the evolution of body shape development, we examined seam cell development of four additional Tylenchomorpha species with vermiform or saccate body shapes.
We confirmed the presence of seam cell homologs and their proliferation in H. glycines. This results in the adult female epidermis having approximately 1800 nuclei compared with the 139 nuclei in the primary epidermal syncytium of C. elegans. Similar to C. elegans, we found a significant correlation between H. glycines body volume and the number and ploidy level of epidermal nuclei. While we found that the seam cells also proliferate in the independently evolved saccate nematode Meloidogyne incognita following infection, the division pattern differed substantially from that seen in H. glycines. Interestingly, the close relative of H. glycines, Rotylenchulus reniformis does not undergo extensive seam cell proliferation during its development into a saccate form.
Our data reveal that seam cell proliferation and epidermal nuclear ploidy correlate with growth in H. glycines. Our finding of distinct seam cell division patterns in the independently evolved saccate species M. incognita and H. glycines is suggestive of parallel evolution of saccate forms. The lack of seam cell proliferation in R. reniformis demonstrates that seam cell proliferation and endoreduplication are not strictly required for increased body volume and atypical body shape. We speculate that R. reniformis may serve as an extant transitional model for the evolution of saccate body shape.
KeywordsSoybean cyst nematode Root-knot nematode Reniform nematode Lesion nematode Aphelenchus Pyriform
differential interference contrast
transmission electron microscopy
How do body shapes evolve? Most nematodes are vermiform (i.e., worm-shaped) throughout post-embryonic development. However, several diverse nematode species develop from vermiform juveniles into saccate adult females. For example, females of the avian parasitic Tetrameridae family have a distended shape . Similarly, the shark-parasitic nematode Phlyctainophora squali develops into a coiled and globose female . Among Tylenchomorpha nematodes, several of the most economically damaging plant-parasitic nematode species develop into saccate-shaped adult females following infection . Heterodera glycines hatches as a vermiform infective second-stage juvenile (J2). Following infection and the establishment of a feeding site, H. glycines females grow disproportionately greater in width than length, developing into saccate-shaped adults. Male H. glycines also initially grows disproportionally in width following infection. During the final juvenile stage (J4), males remodel into vermiform adults.
Not all saccate-shaped nematodes undergo the same sequence of developmental events as H. glycines. The closely related species Rotylenchulus reniformis hatches as a vermiform J2. Interestingly, R. reniformis does not infect following hatching, but rather molts through subsequent vermiform juvenile stages without feeding . Upon molting into a vermiform adult female, R. reniformis infects a host and subsequently develops into a saccate-shaped female.
The evolution of nematode body size was suggested to be due to changes in seam cell proliferation and endoreduplication of nuclei within the epidermal syncytium [8, 12]. However, a recent report demonstrated that a newly isolated Caenorhabditis species evolved increased length due to increased cytoplasmic volume rather than nuclear number or ploidy . We hypothesized that the H. glycines seam cell lineage has undergone extensive evolutionary changes to grow from a vermiform juvenile to a saccate adult female. To understand the evolution of saccate-shaped nematodes, we examined the development of H. glycines as well as the closely related R. reniformis and the independently evolved saccate species M. incognita. As comparisons, we also examined three Tylenchomorpha species that have typical vermiform shapes throughout development.
The number of H. glycines epidermal nuclei increases following infection
Number of epidermal nuclei in H. glycines
No. of epidermal nuclei
H. glycines has seam cell homologs
H. glycines seam cells undergo extensive proliferation following infection and subsequent molts
H. glycines has an epidermal organization similar to C. elegans
H. glycines male seam cells proliferate less than female seam cells
H. glycines epidermal nuclei are polyploid
H. glycines number and ploidy level of epidermal nuclei correlate with body size
We conducted a regression analysis to determine if the increasing number of epidermal nuclei and ploidy level at each successive molt correlate with the increasing body size of H. glycines. We found a significant correlation between the product of number and ploidy level of epidermal nuclei and the estimated body volume (R2 = 0.86; P < 0.0001) of H. glycines at different developmental stages (Fig. 13b). Our results demonstrate that the growth of H. glycines from a vermiform J2 to a saccate adult female is associated with extensive proliferation of the seam cells and polyploidy.
Divergent division patterns acting on conserved seam cells gave rise to independently evolved saccate body shapes
R. reniformis does not use seam cell proliferation to increase body size
Atypical body shape in nematodes is not strictly associated with the number or ploidy level of epidermal nuclei
Our results suggest that the stem cell-like seam cells are conserved in Tylenchomorpha nematodes. Similar to C. elegans, we found that the lateral ridge of H. glycines comprises elliptical-shaped cells with proliferative capacity. Furthermore, using TEM, we observed characteristic adherens junctions along the lateral ridge. Using light microscopy, we found similar cells in all nematodes examined here. While we did not conduct TEM on other species, an examination of published and unpublished TEM data suggests that adherens junctions along the lateral ridge are also present in Meloidogyne incognita and Pratylenchus penetrans [18, 19]. Seam cells are present in bacterial-feeding nematodes Panagrellus redivivus and Pristionchus pacificus [20, 21]. Similarly, the seam cells were found in a group of diverse bacterial-feeding nematodes . We hypothesize that seam cells are conserved within the crown clades of Chromadoria [5, 6].
We show that epidermal proliferation is extensive in H. glycines. Most C. elegans seam cell divisions lead to the formation of a single hyp7 epidermal nucleus and a posterior seam cell . In contrast, we found that seam cell proliferation in H. glycines leads to increasing numbers of epidermal nuclei daughter cells at each successive molt (Fig. 10). Occasionally, the C. elegans seam lineage leads to the formation of glia or neuronal daughters. For example, the C. elegans seam cell V5 leads to the formation of the post-deirid neurons prior to the L2 molt. Due to the parasitic nature of H. glycines development, we were unable to follow the precise lineage of seam cells; however, we did not observe DAPI-stained nuclei with a neuron-like morphology and a position suggestive of post-deirids.
Our results may suggest that epidermal proliferation and the increase in epidermal ploidy cause the increase in body volume of H. glycines. We found a strong correlation between the product of epidermal nuclear number by ploidy and the body volume. Interestingly, we also found that the marked sexual dimorphism in H. glycines is represented by a different number of epidermal nuclei found during the J3 stage. These results support a possible causative link between epidermal nuclei and body size. The primary control strategy for H. glycines is host resistance. H. glycines females are smaller on resistant varieties than on susceptible varieties . It will be interesting to know if this size difference is due to decreased seam cell proliferation or epidermal ploidy.
A question our data still do not answer is why epidermal proliferation in H. glycines does not simply lead to large vermiform nematodes that grow proportionally in length and width? In C. elegans, the elongation of embryos beyond the twofold stage requires the contraction of body wall muscles . Ablation of C. elegans body wall muscle leads to swelling around the destroyed cells . We recently demonstrated that H. glycines and M. incognita undergo progressive muscle atrophy with the onset of the sedentary life cycle . This atrophy corresponds with the increase in size. We speculate that in H. glycines and M. incognita, the atrophy of body wall muscles following infection prevents the maintenance of a vermiform shape by eliminating the lengthening forces found in other nematodes.
Body size in nematodes was hypothesized to have evolved due to changes in the number and ploidy of epidermal nuclei . Our data do not strictly fit that model. While we found a significant correlation between the product of the number and ploidy of epidermal nuclei and H. glycines body size during post-embryonic development; our H. glycines data lay outside of the 95% prediction interval proposed by Flemming et al. . Indeed, based on the number and ploidy of epidermal nuclei in adult H. glycines females, the Flemming model predicts a body volume 100-fold greater than our experimental measurements. We previously showed that the H. glycines epidermis increases in thickness following infection . Also, post-infective H. glycines secretes exudates from the cuticle that are proposed to originate from the thickened epidermis . Perhaps, rather than increased body volume, the unexpectedly large number of epidermal nuclei and increased ploidy in H. glycines are used to provide increased epidermal thickness and protein synthesis.
Our results suggest that multiple mechanisms may lead to a saccate body shape. Similar to H. glycines, M. incognita undergoes extensive epidermal proliferation and epidermal polyploidy. However, the apparent organization of seam cells differs from that observed in H. glycines. The presence of phalloidin-stained structures similar to seam cell apical junctions in the subventral and subdorsal quadrants of M. incognita in post-infective stages could point to the formation of a separate seam cell pool. This will require further examination using both additional light microscopy and TEM. One obvious caveat to the phalloidin staining is its unspecific nature as a general F-actin probe. The unusual pattern of staining in M. incognita may not indicate apical junctions, but rather other actin-enriched structures. As H. glycines and M. incognita likely evolved saccate morphologies independently, we suggest that this serves as an example of parallel evolution where a similar parasitic environment led to selection for increased proliferation of the seam cells acting through distinct mechanisms.
The evolution from vermiform to saccate nematode may have occurred through an intermediate transitional stage as represented by R. reniformis, which shows no obvious epidermal proliferation following infection. The increase in R. reniformis body volume from vermiform to saccate is substantially smaller than that found in H. glycines. H. glycines produces more offspring than R. reniformis [27, 28], and it is generally accepted that saccate nematodes have a greater reproductive potential than vermiform species. We observed that the swelling of R. reniformis began proximal to the growing gonad. In contrast, the increase in H. glycines body volume far outpaces the development of its gonad (here and ). We speculate that the increased fecundity of saccate nematodes began with the evolution of a larger gonad size similar to R. reniformis. This was followed by proliferation of the epidermis allowing for larger gonads with additional reproductive capacity. This hypothesis will require extensive examination of saccate species from additional clades.
Our data reveal that seam cell development has undergone extensive evolutionary changes from C. elegans and that the number and ploidy of H. glycines epidermal nuclei correlate with growth in size. Interestingly, the change in the number of H. glycines epidermal nuclei and ploidy level does not fit a previously established model of body volume in vermiform species, suggesting that these factors may regulate other aspects of development in the Tylenchomorpha species. Our finding of different seam cell division patterns in the independently evolved saccate species M. incognita and H. glycines provides an example of parallel evolution acting through homologous cells. The number of epidermal nuclei and ploidy level is associated with H. glycines and M. incognita shape changes. However, R. reniformis does not demonstrate extensive seam cell proliferation following infection suggesting distinct mechanisms evolved to produce a similar phenotype from a common ancestor. We hypothesize that R. reniformis may serve as an extant transitional model for the evolution of atypical body shapes.
Materials and methods
Heterodera glycines was isolated from a soybean field in Illinois, USA, and maintained on susceptible soybean (cv. Macon) in the greenhouse. To collect synchronized developmental stages of H. glycines, soybean seeds were germinated in moist paper towels for 3 days. Soybean seedlings were then placed in pluronic gel F-127 with freshly hatched J2 for 24 h . Soybean roots were washed with water to remove nematodes that had not infected roots within 24 h of inoculation. Then, H. glycines-infested roots were planted in sandy loam soil and kept in a growth chamber at 22–24 °C and 12-h light cycle until extraction. After inoculation, the infested roots were grown for 6–7 days to collect J3 s, 8–10 days to collect J4 s, 11–13 days for adult females, and 15 days for adult males . Infested roots were macerated with a hand blender to obtain H. glycines at specific time points. The mixture was poured over stacked 850-, 250-, and 25-μm-pore sieves. The developmental stage and sex of each individual were determined based on overall body size and gonad morphology .
Meloidogyne incognita (gift from Dr. Jason Bond) was originally isolated from soybean and maintained on tomato (cv. Rutgers) in the greenhouse. To collect synchronized developmental stages of M. incognita, two-week-old tomato seedlings were inoculated with freshly hatched J2 s. Tomato seedlings were gently removed from the soil and washed with water to remove nematodes that had not infected roots within 48 h following inoculation. M. incognita-infested roots were then planted in sandy loam soil and kept in a growth chamber at 22–24 °C and a 12-h light cycle until extraction. The nematodes were extracted using a hand blender to collect different developmental time points of post-infective J2 s .
Rotylenchulus reniformis (gift from Dr. Martin Wubben) was maintained in sandy loam soil on soybean (cv. Macon) in the greenhouse. To study the non-parasitic stages, eggs were incubated at 30 °C. Different developmental stages were identified based on their morphology and number of cuticles . R. reniformis parasitic adult females were extracted as described above for H. glycines.
Aphelenchus avenae was originally isolated from the rhizosphere of garlic plants and identified using morphological characters. A. avenae was cultured on 1/8 strength Potato Dextrose Agar (PDA) with the fungus Botrytis cinerea . To collect the synchronized A. avenae eggs, gravid females were incubated in 5% M9 buffer . After 5 h, females were removed, and eggs were incubated at room temperature for approximately 48 h. Hatched J2 s were transferred to 1/8 strength PDA with B. cinerea. Developmental stages were determined based on body size and gonad morphology.
Pratylenchus penetrans (gift from Dr. Terry Niblack) was cultured on corn root explants on Murashige and Skoog (MS) media . Specific developmental stages of P. penetrans were isolated by synchronizing populations from eggs. P. penetrans eggs were extracted as previously described . Freshly hatched J2 s were collected and placed on corn root explants on MS media. To recover different developmental stages, P. penetrans culture plates were flooded with water and animals were handpicked under a dissecting scope. Developmental stages of P. penetrans were determined based on body size and gonad morphology.
Helicotylenchus sp. was isolated from soybeans at the University of Illinois research farm and maintained on soybean (cv. Macon) in sandy loam soil in the greenhouse. J4 molting stage animals were determined based on the presence of a shedding cuticle, body size, and gonad morphology.
Live nematode imaging
Nematodes were mounted on a 4% agarose pad on microscope slides kept at room temperature in a dark humidity chamber when not imaging. Slides were rehydrated as needed. Images were taken every hour for 2 days, using an upright compound microscope with a mechanized stage (Zeiss M2 AxioImager and Zen software). Seam cells were identified based on their location and morphology .
DAPI (4′, 6-diamidino-2-phenylindole) staining
Heterodera glycines was fixed in Carnoy’s fixative (60% ethanol, 30% acetic acid, 10% chloroform) overnight at room temperature [8, 17]. Nematodes were transferred to 75% ethanol and stained with 0.2–0.5 μg/ml of DAPI overnight in the dark at room temperature. Images were taken using a Zeiss M2 AxioImager with differential interference contrast (DIC) and fluorescence optics. In C. elegans, the epidermal nuclei are large and flat, contain large nucleoli and are located in four main cords . The number of epidermal-like nuclei was counted from the metacarpus to anus. The number of epidermal nuclei counted on one side of the animal was doubled to estimate the total number of epidermal nuclei. To estimate the body volume of H. glycines, the length of the animal was measured from head to tail, and the mean diameter was calculated from three separate measurements (close to the head, the middle part of the body, and close to the tail) using FIJI. The volume was then estimated based on the equation for the volume of a cylinder during J2 and J3 and the shape of a sphere in adult females. At least six animals were examined for each developmental stage. The data were log transformed and regression analysis was performed using GraphPad Prism version 7.00 for Windows, GraphPad Software, La Jolla California USA, www.graphpad.com.
The ploidy of H. glycines epidermal nuclei was calculated based on previous methods [8, 17]. DAPI-stained sperm nuclei from adult males were used as a haploid control and included in each experimental session. Fluorescence intensity, exposure time, and all other microscope settings were kept consistent during imaging. The fluorescence level of the epidermal nuclei and sperm nuclei were measured using ImageJ software. The nuclei of interest were marked and their integrated density (fluorescence in the area of the region of interest × the mean fluorescence of the region of interest) was measured. The corrected fluorescence intensity was calculated by subtracting the background fluorescence in eight male animals, eight J2 molting stages animals, five J3 females, and 13 adult females . In each animal, the fluorescence intensity of ten epidermal nuclei and ten neuronal nuclei was measured.
All other nematodes (M. incognita, R. reniformis, A. avenae, P. penetrans, and Helicotylenchus sp.) were fixed in Carnoy’s fixative for 2 h and then transferred to 50% methanol and stained with 0.2–0.5 μg/ml of DAPI overnight in the dark at room temperature until imaging. All images were captured with Zen software on a Zeiss M2 AxioImager with DIC and fluorescence optics and analyzed in FIJI. To examine the correlation between body volume and the number of epidermal nuclei among different nematode species, young adult females of M. incognita and R. reniformis, and J4 molting females of A. avenae, P. penetrans and Helicotylenchus sp. were examined. Nematode body shapes at these stages were considered to be cylindrical and the volume calculated as described for H. glycines. At least five animals were examined in each species. The data was log transformed. Regression analysis was performed using GraphPad Prism. Unlike Flemming et al. , we present the number of nuclei as the estimated total number rather than that found on one side of the body. We doubled the number of epidermal nuclei from Flemming et al.  to estimate the total number of nuclei in each species and reproduced the regression model based on their data. We mapped Tylenchomorpha nematode species on their regression model with 95% prediction intervals.
Nematodes were fixed in 4% paraformaldehyde overnight at 4 °C and washed three times with water. Nematodes were placed in a 3-mm Petri dish and cut open with a 1.2 mm × 25 mm BD precision glide needle (Becton, Dickinson, and Company). Nematodes were then transferred to phalloidin (5–7 unit/ml; Thermo Fisher Scientific) and DAPI (0.2–0.5 μg/ml; Thermo Fisher Scientific) and incubated overnight in the dark at room temperature. The images were taken using a confocal microscope (Zeiss LSM 880 Airyscan) and analyzed using ImageJ software.
Synchronized developmental stages of H. glycines were collected from soybean roots and stored overnight at 4 °C. High-pressure freezing and freeze substitution were modified from previous methods used for C. elegans [37, 38]. Metal specimen carriers were coated with 1-hexadecene and a layer of E. coli strain OP50. Nematodes were loaded into carriers with 20% bovine serum albumin and frozen in an HPM 010 high-pressure freezer. Freeze substitution was performed with 2% OsO4 (Electron Microscopy Sciences), 0.1% uranyl acetate (Polysciences), and 2% H2O in acetone in an FS-8500 freeze substitution system. Samples were kept at − 90 °C for 110 h before being warmed to − 20 °C over 5 h. Samples were then kept at − 20 °C for 16 h before being warmed to 0 °C over 5 h. Samples were washed four times in pre-chilled 100% acetone at 0 °C. The last wash was 1 h. Samples were then transferred to room temperature and washed two times in 100% acetone. Samples were infiltrated with 1:1 Polybed812 (Polysciences) resin/acetone for 24 h, 2:1 resin/acetone for 36 h, 100% resin for 24 h, and then changed to fresh resin for 3 days. All infiltration steps were conducted on an orbital shaker at room temperature. Samples were then submerged into embedding molds with resin and hardener and baked at 60 °C for 2 days. 70 nm sections were collected using a PowerTome PC ultramicrotome with a diamond knife and collected onto formvar-coated copper slot grids. Sections were stained with lead citrate and uranyl acetate and imaged with a Phillips CM200 TEM .
ST and NES conceived the project and designed experiments. ST and MK conducted experiments. URC conducted TEM imaging. RJA conducted confocal imaging. ST and NES analyzed results. ST and NES wrote the paper. All authors read and approved the final manuscript.
ST and NES thank Dr. Jason Bond, Dr. Martin Wubben and Dr. Terry Niblack for providing nematodes; Dr. Carrie Butts-Wilmsmeyer for statistical advice; Scott Robinson and the Beckman Institute Imaging Technology Group for assistance with TEM; Dr. Lynn Carta for providing access to Dr. Burt Endo’s original TEM micrographs. ST would like to thank the Schlumberger Foundation, for a Faculty for the Future predoctoral fellowship.
The authors declare that they have no competing interests.
Availability of data and materials
All data generated or analyzed during this study are included in this published article.
Consent for publications
Ethics approval and consent to participate
This research was supported by the Schlumberger Foundation C3560 (076510–02) to ST and NES, the USDA NIFA Hatch program (ILLU-802-934), and NIH NIGMS (R01GM111566) to NES.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 7.Triantaphyllou AC, Hirschmann H. Post infection development of Meloidogyne incognita Chitwood 1949 (Nematoda-Heteroderidae). Annales de l’Institut Phytopathologique Benaki. 1960;1(3):1–11.Google Scholar
- 18.Endo BY, Zunke U, Wergin WP. Ultrastructure of the lesion nematode, Pratylenchus penetrans (Nemata: Pratylenchidae). J Helminthol Soc Wash (USA). 1997;64(1):59–95.Google Scholar
- 19.Endo BY, Wergin WP. Ultrastructure of the second-stage juvenile of the root-knot nematode, Meloidogyne incognita. Proc Helminthol Soc Wash. 1988;55(2):286–316.Google Scholar
- 26.Endo BY. Ultrastructure of cuticular exudates and related cuticular changes on juveniles in Heterodera glycines. J Helminthol Soc Wash (USA). 1993;60:76–88.Google Scholar
- 27.Robinson AF, Inserra RN, Caswell-Chen EP, Vovlas N, Troccoli A. Rotylenchulus species: identification, distribution, host ranges, and crop plant resistance. Nematropica. 1997;27(2):127–80.Google Scholar
- 29.Raski DJ. The life history and morphology of the sugar-beet nematode, Heterodera schachtii Schmidt. Phytopathology. 1950;40(2):135–52.Google Scholar
- 32.De Soyza K. Energetics of Aphelenchus avenae in monoxenic culture. Proceedings of the Helminthological Society of Washington. 1973;40:1.Google Scholar
- 33.Stiernagle T. Maintenance of C. elegans (February 11, 2006), WormBook, ed. The C. elegans Research Community, WormBook, doi/10.1895/wormbook. 1.101. 1.Google Scholar
- 36.Fitzpatrick M. Measuring cell fluorescence using ImajeJ. 2014. https://theolb.readthedocs.io/en/latest/imaging/measuring-cell-fluorescence-using-imagej.html.
- 38.Manning L, Richmond J. High-pressure freeze and freeze substitution electron microscopy in C. elegans. In: Biron D, Haspel G, editors. Methods in molecular biology. Berlin: Springer; 2015. p. 121–40.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.