Original Article

Genome Size Estimates for Crustaceans Using Feulgen Image Analysis Densitometry of Ethanol-Preserved Tissues Nicholas W. Jeffery, T. Ryan Gregory*

Department of Integrative Biology, University of Guelph, Guelph, Ontario, N1G 2W1, Canada Received 20 March 2014; Accepted 18 July 2014 Grant sponsor: Natural Sciences and Engineering Research Council of Canada (NSERC) Graduate Scholarship; NSERC Discovery Grant. Additional Supporting Information may be found in the online version of this article. *Correspondence to: T. Ryan Gregory, Department of Integrative Biology, University of Guelph, 50 Stone Rd. E., Guelph, Ontario, N1G 2W1 Canada. E-mail: [email protected] Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cyto.a.22516 C 2014 International Society for V

Advancement of Cytometry

Cytometry Part A  00A: 00 00, 2014

 Abstract Crustaceans are enormously diverse both phylogenetically and ecologically, but they remain substantially underrepresented in the existing genome size database. An expansion of this dataset could be facilitated if it were possible to obtain genome size estimates from ethanol-preserved specimens. In this study, two tests were performed in order to assess the reliability of genome size data generated using preserved material. First, the results of estimates based on flash-frozen versus ethanol-preserved material were compared across 37 species of crustaceans that differ widely in genome size. Second, a comparison was made of specimens from a single species that had been stored in ethanol for 1–14 years. In both cases, the use of gill tissue in Feulgen image analysis densitometry proved to be a very viable approach. This finding is of direct relevance to both new studies of field-collected crustaceans as well as potential studies based on existing collections. VC 2014 International Society for Advancement of Cytometry  Key terms Crustacea; C-value; DNA content; nuclei

INTRODUCTION

OVER the past 60 years, genome size estimates have been generated for more than 10,000 species of animals, plants, and fungi (1). It has been clear throughout that time that the amount of DNA contained within the haploid genome varies enormously among taxa, but the causes and consequences of this diversity are still being actively investigated. Indeed, genome size data are increasingly sought after for use in comparative studies that aim to elucidate the mechanisms of change and biological implications (or at least correlates) of genome size diversity. These data are also considered important for the genomics community at large, as genome size is a valuable proxy for repetitive DNA content and provides an estimate of the cost and challenge of sequencing and assembling a potential target genome (2,3). The vast majority of genome size estimates for multicellular species have come from two methods: Feulgen densitometry (now most often based on image analysis) and flow cytometry (1). These techniques differ in their chemical and physical underpinnings—one being based on densitometry, the other on fluorescence—but both are well established and have benefited from the development of best-practice procedures (4–8). In general, Feulgen image analysis densitometry (FIAD) makes use of prepared microscope slides containing a monolayer of nucleated cells (e.g., blood or liver in vertebrates, hemolymph or sperm in arthropods). Flow cytometry, by contrast, typically requires large numbers of nuclei to be free in suspension in a buffer solution. In either case, live/fresh or flash-frozen (e.g., in liquid nitrogen or on dry ice) samples of animal tissue have been much preferred over fixed material because the

Original Article nuclei remain intact and can be prepared and stained more easily. Unfortunately, transporting live specimens or preparing flash-frozen samples in the field can often be difficult. A requirement for live or frozen material also significantly limits the utility of existing collections for obtaining genome size data from previously unstudied species. For this reason, there has been significant interest in developing alternative preservation methods and/or developing procedures that will make it possible to use previously preserved material. For example, it has recently been shown that rapidly-dried leaf tissue can be used for flow cytometric genome size estimation in plants (9–11). Storage in concentrated ethanol is by far the most commonly encountered long-term preservation method, in part because it is compatible with DNA extraction and sequencing. However, while ethanol preservation maintains DNA integrity, it also has a desiccating effect on tissues, which renders them hard and rubbery or brittle and makes it challenging to isolate individual nuclei for genome size estimation. Nonetheless, there are a few examples in which ethanol-preserved material has been used. For example, Yang et al. (12) had some success performing flow cytometry (FCM) on ethanol-preserved oyster tissues by pretreating them in hypotonic potassium chloride and then fixing in 75% EtOH. Ethanol preservation has also been used for genome size estimation in fishes using fixed fin clips after extensive chemical digestion and mincing (e.g., (13)). Likewise, ethanol-preserved sponges have been assessed using both FIAD and FCM (14). FCM appears to be particularly challenging to conduct on ethanol-preserved arthropod samples due to tissue dehydration. Given that arthropods are vastly underrepresented in the Animal Genome Size Database in light of their diversity, finding a solution to this impediment could be very important for improving our understanding of genome size diversity and its evolution. The present study included two tests that demonstrate that ethanol-preserved material can be used for genome size estimation of a diversity of crustaceans across a wide range of DNA contents over potentially long periods of time using Feulgen image analysis densitometry, which can allow for separation of nuclei which may be masked by cell debris when using flow cytometry. Using this approach, it was possible to generate new genome size estimates for several previously unstudied crustacean species.

MATERIAL AND METHODS Two distinct tests were conducted as part of the current assessment of using ethanol-preserved materials for genome size estimation by FIAD. The first involved a comparison of results obtained using ethanol-preserved versus fresh/flashfrozen material from a large number of crustacean species exhibiting a 400-fold range in genome sizes. The second aimed to evaluate the effects of long-term ethanol storage on genome size estimation by comparing samples of a single species that had been stored for periods ranging up to 14 years. Test #1: Frozen Versus Ethanol-Preserved Specimens Across a Wide Range of Genome Sizes In total, 37 species were included in this study. They were chosen in an effort to cover both a wide phylogenetic diversity 2

(representing three classes: Branchiopoda, Malacostraca, and Maxillopoda) and a broad range of genome sizes (overall, 0.12–48.96 pg; Table 1). Species were either provided by various colleagues for the study (see Acknowledgements) or collected in the field and identified using the St. Lawrence Marine Species Identification guide (http://slgo.ca/appguidesp/en/index.html) (15), Lamb and Hanby (16), and Belk (17), then verified using DNA barcoding (18).In the majority of cases, tissues from the same individual were used in order to compare results obtained using fresh versus ethanolpreserved tissue. In cases where the individuals were too small (e.g., Daphnia retrocurva), individuals were taken from the same population and slides were made from the entire specimen. Preliminary work revealed that gill tissue provides a reliable number of nuclei that exhibit consistent staining. This was chosen over other commonly used tissues such as hemolymph or muscle, which provide smaller numbers of nuclei and introduce more variability in staining due to differences in nuclear compaction level. For each species, gill was dissected, patted dry to remove excess water, and part was placed in a plastic cryotube and held at 280 C for the frozen samples, while a similar amount of tissue was placed in 2 ml tubes filled with 95% ethanol. The ethanol in the tubes was changed after 24 and 72 h. The gill tissue in ethanol was stored at room temperature in the dark for one year (with the exceptions noted in Table 1). After the period of prolonged storage in ethanol, slides for FIAD were made on the same date for both flash-frozen frozen and ethanol-preserved gill tissue. This was done using a “freeze-flip” procedure. A small amount of gill tissue was placed in 40% acetic acide on a glass microscope slide and then macerated using dissecting pins until no large pieces remained. Each preparation was covered with a glass coverslip, which was compressed onto the slide using three clothespins, and the slide was then frozen on dry ice. After freezing, the clothespins were removed and the coverslip was “flipped” off using the edge of a razor blade and the slides were immersed in 95% ethanol for 1 min. Slides were then allowed to air dry and were stored in the dark for one week prior to staining. Feulgen staining followed Hardie et al. (4). All slides were fixed in methanol:formalin:acetic acid (85:10:5) overnight, rinsed in tap water and hydrolyzed in 5N HCl for 120 min, followed by a brief rinse in dilute HCl and stained in prepared Schiff reagent for 120 min. This was followed by several rinses in bisulfite solution and deionized water. Each set of 100 slides included one slide of chicken blood (Gallus domesticus¸1C 5 1.25 pg) and one of rainbow trout blood (Oncorhynchus mykiss, 1C 5 2.60 pg) as internal standards. Three slides were prepared per species from three separate individuals for each treatment, except in the cases of Spirontocaris sp., Lebbeus groenlandicus, Homarus americanus, and Stegocephalus inflatus where only 1 specimen was available, though 3 slides were still made per species. A minimum of 25 nuclei per slide was measured. For 31 species, flow cytometry (FCM) was also used to estimate genome size. For FCM, either the thoracic appendages or entire individuals in the case of small-bodied species Densitometry Analysis of Ethanol-Preserved Tissues

Original Article Table 1. Haploid (1C) genome size estimates for all 37 species in this study, including flow cytometry (FCM) of fresh material as well as Feulgen image analysis densitometry (FIAD) of flash-frozen tissue and tissues preserved in 95% ethanol at room temperature for one year

CLASS

Branchiopoda Branchiopoda Branchiopoda Branchiopoda Branchiopoda Maxillopoda Maxillopoda Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca Malacostraca

SPECIES

COMMON NAME

FCM/FLASHFROZEN (PG)

Daphnia retrocurva Triops australiensis Bythotrephes longimanus Artemia salina Eubranchipus bundyi Heterocope septentrionalis Epischura baikalensis Thysanoessa sp. Euphausia frigida Euphausia triacantha Euphausia superba Meganyctiphanes norvegica Mysis sp. Sphaeromatidae Idotea sp. Armadillidium vulgare Homarus americanusa Hemigrapsus nudus Litopenaeus vannameib Spirontocaris sp. Platorchestia platensis Pagurus sp. Pagurus acadianus Pagurus pubescens Lebbeus groenlandicus Pandalus montagui Pandalus borealis Gammarus lacustris Gammarus oceanicus Crangonyx pseudogracilis Gammaracanthus loricatus Micruropus wahli Eulimnogammarus cyaneus Caprella sp. Caprella septentrionalis Stegocephalus inflatus Unknown marine amphipod

Water flea Tadpole shrimp Spiny water flea Brine shrimp Fairy shrimp Copepod Copepod Krill Krill Krill Antarctic krill Northern krill Opossum shrimp Marine isopod Marine isopod Terrestrial isopod American lobster Purple shore crab Whiteleg shrimp Blade shrimp Sand flea Hermit crab Acadian hermit crab Hairy hermit crab Spiny lebbeid Pink shrimp Northern shrimp Freshwater amphipod Marine amphipod Freshwater amphipod Marine amphipod Baikal amphipod Baikal amphipod Skeleton shrimp Skeleton shrimp Marine amphipod Marine amphipod

0.25 0.16 0.27 1.90 1.05 5.49 N/A 12.77 34.93 30.06 48.53 18.75 12.00 4.21 N/A 1.70 3.97 3.45 2.89 11.83 1.84 8.48 7.95 12.17 20.77 13.47 12.69 9.35 11.5 6.68 10.34 N/A N/A N/A 1.54 50.91 N/A

FIAD/FLASHFROZEN (PG)

FIAD/ETHANOLPRESERVED (PG)

RELATIVE DIFFERENCE (%)

ABSOLUTE DIFFERENCE (PG)

0.25 0.12 0.31 1.84 1.10 5.80 0.82 12.52 34.74 27.56 48.96 20.00 10.81 4.19 3.69 1.74 3.13 3.42 2.67 12.86 1.82 7.97 6.635 12.22 21.49 13.66 12.61 8.5 11.83 5.55 10.47 4.12 3.81 0.98 1.34 N/A 3.95

0.23 0.16 0.29 1.69 1.02 5.54 0.84 12.56 34.92 28.38 52.92 23.61 11.94 4.17 3.36 1.88 3.21 3.38 2.61 12.25 1.73 7.47 6.35 10.995 21.06 12.17 12.05 8.58 10.55 5.15 12.17 3.75 3.87 0.86 1.37 50.57 4.24

8.00 33.33 6.45 8.15 7.27 4.48 2.44 0.32 0.52 2.98 8.09 18.05 10.45 0.48 8.94 8.05 2.56 1.17 2.49 4.75 4.95 6.27 4.30 10.02 2.00 10.91 4.44 0.94 10.82 7.21 16.24 8.98 1.57 12.24 2.24 0.67 7.34

0.02 0.04 0.02 0.15 0.08 0.26 0.02 0.04 0.18 0.82 3.96 3.61 1.13 0.02 0.33 0.14 0.08 0.04 0.07 0.61 0.09 0.50 0.285 1.225 0.43 1.49 0.56 0.08 1.28 0.40 1.70 0.37 0.06 0.12 0.03 0.34 0.29

The relative (percent) differences and absolute differences (in picograms) are shown for flash-frozen versus ethanol-preserved samples by FIAD. a Stored in ethanol for 3 years. b Stored in ethanol for 2 weeks. All other ethanol-preserved samples were stored for one year prior to analysis.

were crushed with a glass tissue grinder in 500 ll of cold LB01 buffer (19) and coprepared with either chicken (Gallus domesticus; 1C 5 1.25 pg), rainbow trout (Oncorhynchus mykiss; 1C 5 2.60 pg), or fly (Drosophila melanogaster; 1C 5 0.18 pg) as an internal standard. The genome size of rainbow trout was calculated against the chicken blood, and the values used for both chicken and Drosophila are based on previously reported values. Each sample was treated with 2 ll RNase (4 lg/ml) Cytometry Part A  00A: 00 00, 2014

and stained with 12 ll of propidium iodide (24 lg/ml) on ice in the dark for 1 h. Each sample was run on a Beckman Coulter FC500 flow cytometer using a 488 nm laser (Supporting Information Fig. S1). A minimum of 2,000 nuclei was analyzed for three replicates per species, except in the cases noted above where pseudoreplication was used. The FCM value for S. inflatus is from Rees et al. (20). CVs were always 0.99).

possible to obtain estimates from specimens that have been stored for over a decade. If freezers are unavailable or if long-term storage is not required, then specimens may be preserved in ethanol in the field for anywhere from several days to over a year and still provide genome size estimates that are close to those provided by fresh or frozen samples. This will prove especially useful in remote locations where using liquid nitrogen or dry ice is difficult. It will also permit estimates to be generated from specimens from existing collections, provided that they have been properly preserved and stored. It is possible that this method of preservation will also be suitable for genome size estimation in other arthropods (and perhaps other phyla); however, compared to other arthropods, crustaceans have easily accessible, thin sheets of tissue (the gills), which produce a monolayer of undamaged nuclei when properly prepared. This cannot easily be achieved using other tissues such as muscle or brain, which may make it difficult to implement in terrestrial arthropods. Thus, whereas simple ethanol preservation for genome size estimation appears very promising for crustaceans, follow-up studies of other animal taxa are needed. Wherever possible, type of tissue and approximate genome size should be matched between unknowns and standards. In addition to testing specimen preservation methods, this study provided genome size estimates for 37 species of crustaceans, 29 of which represent new additions to the Animal Genome Size Database. Seven species of shrimp with new estimates had genome sizes >10 pg, which qualifies them as “large” based on the classification of Dufresne and Jeffery (21). Of the eight species with previously estimated genome sizes, the current estimates from both frozen and ethanolpreserved specimens are in generally good agreement with them, with the exception of A. salina where the previous estimate of 2.91 pg is considerably larger than our estimate of 1.90 pg. It is worth noting, however, that different populations of this brine shrimp exhibit different ploidy levels, which could account for this difference (22,23). By contrast, the 5

Original Article

Figure 3. Examples of Feulgen-stained gill tissue nuclei from specimens of Gammarus lacustris that had been collected and stored in ethanol in different years: (A) 1997, (B) 2001, (C) 2005. Panel (D) shows damaged nuclei that resulted from improper preservation methods (sample from 2007). Images taken at 1003 magnification. Scale bar equals 20 lm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

present estimates for L. vannamei, P. platensis, and S. inflatus were all

Genome size estimates for crustaceans using Feulgen image analysis densitometry of ethanol-preserved tissues.

Crustaceans are enormously diverse both phylogenetically and ecologically, but they remain substantially underrepresented in the existing genome size ...
323KB Sizes 4 Downloads 5 Views