Nitrogen excretion during marsupial development in the terrestrial isopod Armadillidium vulgare

Nitrogen excretion during marsupial development in the terrestrial isopod Armadillidium vulgare

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99 Contents lists available at ScienceDirect Comparative Biochemistry and Physiology, ...

942KB Sizes 1 Downloads 62 Views

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part A journal homepage: www.elsevier.com/locate/cbpa

Nitrogen excretion during marsupial development in the terrestrial isopod Armadillidium vulgare Emily Rockhill, Kenna Schammel, Alfredo Reyes-Guzman, Zechariah Harris, Jonathan Wright

T



Department of Biology, Pomona College, 175 West 6th Street, Claremont, CA 91711, USA

ARTICLE INFO

ABSTRACT

Keywords: Ammonia Embryo development Isopod Marsupium Nitrogen Uric acid

Marsupial embryos of Armadillidium vulgare (Isopoda: Oniscidea) were collected at different stages of development and assayed for products of nitrogen excretion. Stages were classified as early stage one, late stage one (clear embryo and somite differentiation), early stage two (chorion shed, prior to blastokinesis), late stage two (following blastokinesis), and mancae (vitelline membrane shed; second embryonic molt). Stage one and stage two embryos were primarily ammonotelic. Mancae showed a significant increase in stored uric acid and decrease in ammonia production, in most cases to undetectable levels. The increased metabolic rate of mancae, and the fact that they imbibe marsupial fluid prior to exiting the marsupium, may have favored a switch from ammonotely to uricotely to avoid ammonia toxicity. Protein metabolism, estimated from ammonia production, accounted for 7% of the measured catabolic rate in Stage 2 embryos. Newly emerged juveniles showed a > 2-fold increase in metabolism relative to mancae, accompanying the transition from aquatic to aerial respiration. Following 48 h post-emergence, juveniles resumed ammonia excretion, volatilizing the base (NH3) as in later instars. Elevated ammonia excretion in early juveniles may derive from the catabolism of remaining yolk protein. A sharp increase in whole-animal glutamine in juveniles is consistent with its role as an intermediary nitrogen store during periodic ammonia excretion. Total ammonia concentration in the marsupial fluid fluctuated but did not increase significantly over time and ammonia was not volatilized across the oostegites, indicating that embryo ammonia is transported into the maternal hemolymph for excretion.

1. Introduction

marsupium in the Oniscidea has been studied by Hoese (1984) and Hoese and Janssen (1989). Following fertilization, female isopods undergo a specialized parturial molt and develop leaf-like extensions of the coxae, the oostegites, on the first 5 pairs of the pereopods. The oostegites overlap in an imbricate arrangement and form the floor of the marsupium. In marine isopods, the marsupium opens to the anterior and the embryos are perfused directly with seawater (Hoese, 1984). In the Oniscidea, the anterior and posterior oostegites insert into grooves on the respective sternites. Members of the basal section Ligiamorpha possess an open or amphibian type marsupium in which the marsupial fluid is provisioned externally via the pleural water capillary system; rows of small setae on the 6th and 7th pereopods form a capillary channel when these are appressed, serving in water uptake (Hoese, 1984, 1981). Water is sourced either from seawater (most Ligia species) or from freshwater (Ligidium spp.) supplemented with maternal ions (Yoshizawa and Wright, 2011). In the Holoverticata, comprising the

The transition from planktotrophic development in larval crustaceans to direct development via brooded, lecithotrophic eggs is seen in several independently derived lineages and is probably one of the more significant preadaptations enabling the broad adaptive radiations of the Amphipoda and Isopoda in the intertidal. In these two orders, uniquely, non-insect crustaceans have attained a fully terrestrial habit. The terrestrial isopods of the suborder Oniscidea represent a major radiation with over 3600 described species (Schmalfuss, 2003; Schmidt, 2008). Many of these are mesic-xeric species and a few, including Porcellio brevicaudatus of the Negev Desert (Kashani et al., 2011; Shachak and Yair, 1984), and Venezillo arizonicus of the Mojave and Sonoran-Colorado Desert (Warburg, 1965a, 1965b), exploit some of the hottest and driest habitats on Earth. The comparative anatomy of the maternal brood pouch or

Abbreviations: ES1, early stage 1 embryos; LS1, late stage 1 embryos; ES2, early stage 2 embryos; LS2, late stage 2 embryos; EM, early mancae; LM, late mancae; VCO2, CO2 flux (nl min−1 or ml min−1) ⁎ Corresponding author. E-mail addresses: [email protected] (E. Rockhill), [email protected] (K. Schammel), [email protected] (A. Reyes-Guzman), [email protected] (Z. Harris), [email protected] (J. Wright). https://doi.org/10.1016/j.cbpa.2018.09.029 Received 26 September 2018; Accepted 26 September 2018 Available online 05 October 2018 1095-6433/ © 2018 Elsevier Inc. All rights reserved.

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al.

Fig. 1. Marsupial embryo stages of A. vulgare. Top row, left to right: early stage 1 (ES1), late stage 1 (LS1), early stage 2 (ES2). Bottom row: late stage 2 (LS2), manca. The chorion is shed as a thin, glassy envelope between LS1 and ES2. In ES2, the yolk mass is assimilated into the presumptive midgut followed by blastokinesis which marks the transition to LS2. Shedding of the vitelline membrane from LS2 yields the mancae.

accumulation. Gravid females may possess a mechanism for excreting marsupial fluid ammonia, for example volatilizing NH3 across the oostegites or transporting ammonia into the maternal hemolymph and volatilizing NH3 from the pleopodal fluid in the usual manner (Wieser et al., 1969; Wieser and Schweizer, 1970; Wright and O'Donnell, 1994). Alternatively, embryos may switch to storage excretion during all or part of marsupial development, as in the cleidoic eggs of reptiles and birds. Possible candidates for storage excretion include urea, purines and nitrogen-rich amino acids such as glutamine. Uric acid is present in small amounts (typically 1–3 μmol g−1) in probably all Oniscidea (Linton et al., 2016), and all crustaceans appear to possess xanthine oxidoreductase and other critical enzymes for urate synthesis (Hartenstein, 1968; Linton et al., 2016). Glutamine is known to serve as a temporary nitrogen store in terrestrial (Wieser and Schweizer, 1972; Wright et al., 1994; Wright and Peña-Peralta, 2005) as well as littoral (Nakamura and Wright, 2013) Oniscidea.

sections Tylomorpha, Synocheta, Mesoniscidea and Crinocheta (see Schmidt, 2008), the marsupium is provisioned with fluid from the maternal hemolymph, and in the Crinocheta this employs modified segmental cotyledons (Akahira, 1956; Hoese and Janssen, 1989); embryo development in these sections is thus independent of external liquid water. When they initially pass from the ovary into the marsupium, oniscidean eggs possess two extra-embryonic membranes, an outer chorion and inner vitelline membrane (Strömberg, 1965, 1972), and are filled with vitellocytes. As the embryo develops, the yolk mass decreases in volume and energy reserves are assimilated into the presumptive midgut (Fig. 1). After 10–12 days, eggs of Armadillidium vulgare shed the chorion at the first embryonic molt and transition from Stage 1 to Stage 2 (Surbida and Wright, 2001). After a further 5–6 days, the yolk mass disappears and the embryo undergoes blastokinesis, rotating about the longitudinal axis, so the ventral surface and pereopods now orient to the concave face of the egg. Blastokinesis coincides with the atrophy of the embryonic dorsal organ (Goodrich, 1939; Wright and O'Donnell, 2010). After the subsequent shedding of the vitelline membrane, Late Stage 2 embryos become the first instars or mancae which remain in the marsupium for a further 5–6 days. During this period, they imbibe the marsupial fluid (Hoese and Janssen, 1989; Surbida and Wright, 2001) before emerging as free-living juveniles. Energy reserves in the yolk include both vitellin proteins and lipids (Okuno et al., 2000). The yolk proteins provide an important catabolic substrate as well as serving for anabolic tissue growth (Rønnestad et al., 1999). Very little is known, however, about waste nitrogen excretion by the embryos of crustaceans with lecithotrophic development. Crustaceans generally eliminate deaminated nitrogen from amino acid catabolism as ammonia, either via the antennal gland (Binns, 1969; Greenway, 1991) or the gill epithelia (Weirauch et al., 2004). The one notable exception is the coconut crab Birgus latro which is purinotelic, synthesizing a mixture of guanine and urate in the fat body (Greenaway and Morris, 1989; Linton et al., 2016); purines are transported across the midgut and voided as a white pellet. Among other groups with lecithotrophic eggs, precipitated purines comprise about 50% of the accumulated waste nitrogen stored in the allantoic sac of the American alligator Alligator mississippiensis (Clark et al., 1957), while other reptile and bird eggs mostly accumulate urea (Fisher and Eakin, 1957; Packard and Packard, 1983, 1987; Sartori et al., 2012). By contrast, the aquatic, yolky eggs of fish (Wright and Fyhn, 2001), amphibians (Munro, 1953), and freshwater snails (Sloan, 1964), are primarily ammonotelic. The present study sought to examine nitrogen excretion during and immediately following marsupial development in Armadillidium vulgare (Oniscidea, Crinocheta, Armadillidiidae). Embryo excretion in the Oniscidea represents an interesting case. Being aquatic, embryos could release ammonia directly into the marsupial fluid, but the limited fluid volume of the marsupium imposes the potential for toxic ammonia

2. Materials and methods 2.1. Collection and culture Armadillidium vulgare were collected from the campus of Pomona College and in the local San Gabriel Mountains and maintained in lab terraria with oak litter and occasional potato and carrot as supplementary food. Gravid females were identified by the tumid ventral marsupium, and it was typically possible to identify the brood stage using x40–80 magnification. Suitable embryos were sampled by lifting the anterior edges of the oostegites gently using fine forceps and transferred to marsupial fluid saline solution. The saline was based on previously measured electrolyte concentrations (Yoshizawa and Wright, 2011): 280 mM NaCl, 23 mM KCl, 13 mM CaCl2, 5 mM MgSO4, and 5 mM HEPES buffer, adjusted to pH 7.8. Although prior work (Ban, 1950; Surbida and Wright, 2001) has shown that embryos can be reared from Stage 1 through to the manca stage in a similar saline solution, all experiments here were initiated within 1 h of isolating embryos from the female. 2.2. Staging embryos Embryos were categorized into five developmental stages following Surbida and Wright (2001), (Fig. 1): Early Stage 1 (ES1), Late Stage 1 (LS1), Early Stage 2 (ES2), Late Stage 2 (LS2), and mancae. ES1 possess minimal embryo differentiation while in LS1 the embryo somites are clearly visible. Following shedding of the chorion, the embryos are classified as Stage 2. ES2 and LS2 are pre- and post-blastokinesis respectively. 93

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al.

2.3. Ammonia assays

2.7. Embryo metabolism

Excretion of diffusible ammonia was studied by using a pipettor to sample a known number of embryos following isolation into marsupial fluid saline; pipette tips were cut back a short distance to allow uptake of the eggs. The sample was first expelled onto a dry glass surface with minimal saline, and the remaining liquid removed. The pipettor was then used to collect 40 μl of clean saline which served to suspend and resample the eggs. The 40 μl droplet containing the eggs was expelled into a 1 ml (20 × 20 mm) watch-glass which was covered with a 22 × 22 mm cover glass and sealed with vacuum grease. After 24 h, a 2 μl sample of the droplet was collected and assayed for ammonia using a scaled-down version of the Sigma procedure AA0100 (Sigma-Aldrich, St. Louis, MO). Absorbances at 340 nm were measured using an Implen NanoPhotometer P-Class (Implen Inc., Westlake Village, CA). This procedure was modified slightly to measure possible gaseous ammonia excretion in juveniles. The cut-off lid of a 1.5 ml Eppendorf microcentrifuge tube was placed inverted in the watch-glass and filled with 60 μl of 0.1 M HCl to make an acid trap. Juveniles were collected either directly from cultures, and massed, or were collected following their release from isolated gravid females to allow precise dating of emergence times. Small numbers of animals (3−10) were used in individual experiments and placed in the watch glass which was sealed as before. This design left a 1 mm gap between the top of the acid trap and the cover glass and successfully prevented animals accessing the acid trap. Volatile ammonia collected in the acid trap was assayed in a 2 μl sample after 24 h. Two types of controls were conducted with the ammonia assays. Assay recovery was tested by measuring the absorbance of a known (588 μM) control solution and gave a mean recovery of 101.5 ± 6.0% (n = 9). Potential leakage of NH3 from the watch glass was tested by substituting the egg sample with a 50 μl droplet of 3 mM or 600 μM ammonium chloride then assaying the droplet after 24 h; these yielded respective recovery values of 98.2 ± 2.6% (n = 8) and 98.5 ± 12.3% (n = 6).

In order to estimate the fractional contribution of protein catabolism to the total embryo catabolism at different stages of development, we measured the VCO2 of embryos using stop-flow respirometry. Staged embryos were isolated in 50 μl of saline and the saline droplet expelled into light mineral oil. A minimal volume of oil containing the saline droplet was then collected using a pipettor and expelled into a 10 ml glass respirometry chamber. The chamber was connected to an air pump and the air vented into desiccant (Drierite™) before passing into a Sable Systems CA-10 CO2 analyzer (Sable Systems Inc., Henderson, NV) and flow meter. Digitized data output was recorded and analyzed using Sable Systems Expedata software. For stop-flow measurements, the chamber was flushed with CO2-free air for 10 min to purge dissolved CO2 from the oil, and then isolated using tubing valves while the airflow was diverted through a shunt pathway. After allowing embryo CO2 to accumulate for 30 min, the chamber was flushed again, resulting in the release of a discrete burst of CO2. 2.8. Test of possible ammonia diffusion across the oostegites Ammonia released by marsupial embryos could potentially be volatilized across the oostegites. To examine this possibility, marsupial females (n = 10) were placed, inverted, within a small corral constructed of Fun-Tak™ on the inverted lid of a 2 ml plastic Petri dish, and constrained by application of the base. Two narrow (2 × 10 mm) strips of filter paper were taped to the underside of the base, moistened with 5 μl of 1 mM bromothymol blue, and aligned so as to overly the marsupium and pleon. The pH-dependent color change from orange to blue provides a sensitive indicator of alkalination resulting from volatilized ammonia (Hunter and Uglow, 1993). 2.9. Marsupial fluid ammonia and volume measurements To examine the effects of embryo ammonia excretion on marsupial fluid ammonia levels, we measured marsupial fluid volume and ammonia concentrations during the different marsupial brood stages. Ammonia concentrations were determined for small (0.2–1.0 μl) volumes of marsupial fluid using the Sigma AA0100 assay as described above. Samples were collected by inserting a pulled glass micropipette between the oostegites and transferred to 1 ul microcapillaries (Drummond, Broomall, PA) for volume determination. To estimate the total marsupial fluid volume, embryos were first removed using fine forceps, transferred to a waxed weigh-paper, and promptly massed. A small piece of tared filter paper was then used to blot the remaining marsupial fluid from the marsupium and the volume again determined gravimetrically.

2.4. Urea Assays were conducted to quantify potential excretion of urea into saline, as well as accumulation in the embryo. Diffusible urea was studied by transferring embryos to 40 μl saline droplets in watch glasses as for diffusible ammonia. For whole-embryo urea assays, embryos were homogenized in 50 μl of ice-cold 1.7 M perchloric acid to precipitate protein, sonicated, then after 2 min diluted with 350 μl deionized water and neutralized with 22 μl of a solution of 4 M KOH and 0.4 M KCl. The neutralized sample was centrifuged for 10 min at 5000g and the supernatant assayed for urea using the Sigma assay MAK006, determining the concentration against a standard curve. Spiked controls were used to quantify the assay recovery.

2.10. Statistical analysis Data were analyzed using MS Excel and SPSS Version for Windows. Comparisons across groups used a 1-way ANOVA and comparisons between means used 2-sample t-tests assuming unequal variance. Data are presented as mean ± SEM and statistical significance assumes α = 0.05 throughout.

2.5. Uric acid For urate assays, embryos were homogenized in saline, deproteinized with 1.7 M ice-cold perchloric acid as for urea, sonicated, neutralized and centrifuged. Uric acid was measured in the supernatant using Biovision (Biovision Inc., Milpitas, CA) or Sigma MAK077 assays, determining the concentration from a standard curve. Spiked controls were used to quantify assay recovery.

3. Results 3.1. Ammonia excretion

2.6. Glutamine

Diffusible ammonia excretion rates for the different embryo stages measured over a 24-h period are shown in Fig. 2. Ammonia levels, here and elsewhere, refer to ‘total ammonia’ (NH3 + NH4+). Ammonia excretion increased significantly between LS1 and ES2, and then decreased significantly between LS2 and mancae. In 7 of the 9 manca broods assayed, diffusible ammonia excretion was unmeasurable. As

Glutamine was determined using a scaled-down version of Sigma procedure GLN1. Embryos from individual broods were staged, deproteinized, sonicated and centrifuged as for uric acid, and the supernatant assayed, determining the soluble glutamine from a standard curve. 94

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al.

Table 2 Diffusible urea excretion from embryos. N refers to the number of overnight runs assayed, using 18–45 embryos per run.

Urea (nmol d SEM N

−1

)

Stage 1

Stage 2

Mancae

0.153 0.076 9

0.058 0.052 4

0.057 0.037 4

Fig. 2. Ammonia excretion rates (nmol day−1 embryo−1) for the different brood stages. Ammonia excretion increases during the rapid embryonic development in Stage 2 but decreases sharply in mancae. Asterisks show significant differences with respect to the prior stage (*, p < .05; *** p < .0005). (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2).

Fig. 4. Uric acid content (nmol. Embryo−1) of embryo stages. Mean uric acid content increases sharply in mancae. Bars show mean ± SEM with the number of broods assayed in each case. Some low levels in early mancae indicate that the main period of increased urate storage begins after the second embryonic molt, and accounts for the high variance in the manca contents. ** = significant difference between LS2 and mancae (p = .0037). (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2).

amount. Mass-corrected values (Table 1) show no significant difference across stages (p > .1; single-factor ANOVA). Only trace amounts of diffusible urea were detected from embryos incubated overnight in saline (Table 2). Fig. 3. Urea content of the different marsupial stages. Bars show mean ± SEM. Sample sizes above the error bars show the number of separate assays for each stage, each utilizing 8–40 embryos. (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2).

3.3. Uric acid storage Whole-embryo uric acid contents are shown in Fig. 4. Standard curves for assays had r2 values between 0.93 and 0.99. Controls were performed using samples spiked with 1–8 nmol uric acid prepared in marsupial fluid saline and yielded a recovery of 102 ± 19.7% (n = 10). Measured levels of urate remained low (< 1 nmol embryo−1) until the manca stage when they increased dramatically. The approximately 17-fold increase in whole-embryo urate between late Stage 2 embryos and mancae represents a similar mass-specific increase (Table 3).

mancae consume the marsupial fluid and transition to air-breathing, about 4 days after the second embryonic molt, they become unable to survive overnight immersion in saline. The data shown in Fig. 2 thus comprise only these younger mancae. 3.2. Urea storage and excretion Small amounts of urea were detected in most embryos (Fig. 3), increasing to 0.56 nmol. individual−1 in mancae. Samples spiked with urea recovered 100 ± 23% (mean ± SEM; n = 5) of the spike

Table 3 Mass-corrected uric acid levels in marsupial stages and juveniles (0.8–5 mg). The sharp increase in whole-embryo urate in the manca stage (Fig. 3) is mostly explained by a substantial mass-specific increase. The increase in whole-animal urate continues in juveniles although mass-specific levels show a modest decline. (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2).

Table 1 mass-corrected urea levels (nmoles mg−1) in marsupial stages and juveniles (0.8–1.8 mg). N refers to the number of separate assays performed, each utilizing 8–40 embryos or 4–6 juveniles. Juvenile masses ranged from 1.38–1.98 mg. (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2).

Mean SEM N

−1

ES1

LS1

ES2

LS2

Mancae

Juveniles

nmol. mg

1.71 0.771 8

0.627 0.416 4

1.76 0.263 9

1.84 0.279 8

2.29 0.602 7

0.68 0.08 4

nmol. embryo−1 N

95

Mean SEM Mean SEM

ES1

LS1

ES2

LS2

Mancae

Juveniles

2.72 0.74 0.33 0.089 12

1.74 0.70 0.43 0.18 7

3.32 0.99 0.83 0.25 11

1.10 0.26 0.26 0.070 12

16.4 4.5 4.42 1.2 20

11.0 2.2 16.6 2.9 9

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al.

Fig. 5. Mass-specific ammonia excretion rates (nmol g−1 h−1) in juvenile A. vulgare. Each data point represents the rate determined from a sample of 3–10 animals. Volatilization rates are small or unmeasurable in the first 2 days after juveniles emerge from the marsupium (see text) but increase sharply in the first week. Ammonia production declines thereafter as juveniles increase in mass, stabilizing when they reach 4–5 mg. This corresponds to an age of approximately one month at 20 °C (Helden and Hassall, 1998).

Fig. 7. Catabolic rates measured as VCO2 (nL min−1) for the brood stages of A. vulgare. Bars show ± SEM. Sample sizes show the number of separate trials, each using 15–173 embryos.

individuals of all stages, but increase significantly in juveniles. 3.6. Metabolism

3.4. Juvenile ammonia excretion

Stop-flow measurements of VCO2 (nl min−1) for the different embryo stages were calculated from the product of the integrated CO2 peak (%CO2 min) and air flow rate (ml min−1), then dividing by the period of chamber isolation (min). Results are shown in Fig. 7. Stage 1 embryos averaged 0.67 ± 0.067 nl min−1 (n = 14) and Stage 2 embryos averaged 1.08 ± 0.090 nl min−1 (n = 17). Measured VCO2 increased progressively during the manca stages, reaching 3.03 ± 0.73 nl min−1 (n = 5) in late mancae. Assuming a mean energy stoichiometry of 23 kJ l CO2−1 these represent mean rates of 0.26 μW , (early stage 1), 0.41 μW (stage 2) and 1.16 μW (late mancae). VCO2 was also measured for juvenile and adult A. vulgare, ranging in mass from 0.29 to 207.2 mg. A log-log plot of VCO2 against mass (Fig. 8) shows a tight linear relationship, closely conforming to Kleiber's Law (β = 0.739). The allometric equation describing the curve gives a predicted VCO2 for a 0.3 mg juvenile of 7.2 nl min−1 and the mean measured value for smallest juveniles = 6.7 ± 0.71 nl min−1 (n = 5). The mean mass of these juveniles was 0.31 ± 0.009 mg. The early

Within the first 2 days following emergence from the marsupium, juveniles show negligible volatile ammonia excretion. In the next few days, mass-specific ammonia production increases sharply, reaching 1100–1200 nmol g−1 h−1 within the first 2–3 weeks while juveniles are still < 1 mg in mass (Fig. 5). Thereafter, mass-specific excretion decreases progressively, attaining stable rates of 50–200 nmol g−1 h−1 by the time they reach 4–5 mg. Helden and Hassall (1998) calculated offspring (post-marsupial) growth rates of A. vulgare at 20 °C to be approximately 0.14 mg d−1, making 4–5 mg juveniles about a month old. Growth is, however, strongly influenced by temperature and nutritional status. 3.5. Glutamine storage Whole-animal glutamine levels (nmoles) were determined for all marsupial stages and first-instar juveniles (Fig. 6). Values remain low throughout marsupial development, and were undetectable in several

Fig. 8. Log-log plot of VCO2 (ml min−1) against mass (g) in A. vulgare showing the tight allometric scaling of metabolic rate with body mass ((β = 0.74 ± 0.015); r2 = 0.993) and the sharply lower mass-specific metabolism in late mancae (square symbol; mean ± SEM). N = 19 (juveniles and adults); N = 5 (mancae).

Fig. 6. Glutamine levels (nmoles per individual; mean ± SEM) measured in the marsupial stages and in first instar juveniles. The asterisk denotes a significant increase with respect to the prior stage (p = .013). Mean juvenile mass = 0.69 ± 0.037 mg. (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2; Juv, juveniles). 96

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al.

Marsupial ammonia (mM)

10

Brood Stage ES1 Fractional volume Volume embryo−1 (μl)

0.60 0.232

ES2 0.35 0.113

LS2 0.42 0.133

EM 0.44 0.157

140

7

9

10 120

12 8

100

7

8

6

80

7 5 4

60

3

40

2 20 1 0

0 ES1

juvenile VCO2 is more than twice that of late mancas (3.0 ± 0.73 nl min−1; n = 5), despite their similar mass. The difference is highly significant (p < .005; 2-sample t-test).

Predicted ammonia (mM)

Table 4 Marsupial fluid volumes (fractional volume, and volume (μl) per embryo) measured for gravid females at the different brood stages. Values are means of 2–3 measurements. Neither measure showed significant variation as a function of brood stage (single-factor ANOVA; p > .5). (ES1, early stage 1; ES2, early stage 2; LS2, late stage 2; EM, early mancae).

LS1

ES2

LS2

EM

Marsupial Stage Fig. 9. Marsupial fluid total ammonia concentration measured for the different brood stages (ES1, early stage 1; LS1, late stage 1; ES2, early stage 2; LS2, late stage 2; EM, early mancae); means ± SEM (n = 7–12). The line graph (secondary y-axis) gives the predicted ammonia concentrations at the end of each respective brood stage assuming the measured rates of ammonia excretion and ammonia retention, using the development times given in Surbida and Wright (2001).

3.7. Ammonia excretion in gravid females Qualitative assays examining ammonia volatilization across the ventral surface of females showed localized blue staining (alkalinization) of Bromothymol Blue above the pleopods in nine of ten trials after 24 h. All animals resumed normal activity immediately following removal from the chamber. Blue coloration appeared abruptly, and often multiple times (1–6) over the course of a day. In contrast, no blue coloration appeared over the oostegites in any of the trials.

diffusible urea (< 0.2 nmol d−1), and whole-embryo urea and uric acid levels do not show systematic changes prior to the manca stage (second embryonic molt). Assuming a stoichiometry of 0.8 l CO2 g protein−1 (Schmidt-Nielsen, 1997), and assuming the nitrogen content of yolk vitellins to be similar to that measured for chicken eggs (15%; Calvery and White, 1932), yolk metabolism will consume 0.8/0.15 = 5.3 l CO2 g N−1, or 74 l CO2 mol. N−1. The metabolic rate in Stage 2 embryos (1 nl CO2 min−1 or 1440 nl d−1) would thus require excretion of 1440/ 74 = 19 nmol NH3 d−1 if fueled entirely by protein. The measured ammonia excretion in Stage 2 embryos (1.2–1.4 nmol d−1) indicates that protein metabolism accounts for approximately 7% of embryo catabolism. The remainder is probably fueled by the yolk lipids with the majority of yolk protein being assimilated into tissue growth. The production efficiency for isopod embryos is not known, but typical values of 51–70% for other animal embryos (Withers, 1992) indicate a likely anabolic rate in Stage 2 embryos of 1–1.6× the catabolic rate, or 0.23–0.37 μW. Following blastokinesis, oniscidean embryos enter late Stage 2 and the embryo dorsal organ atrophies (Goodrich, 1939; Fiorini, 1980; Wright and O'Donnell, 2010). Significant ammonia excretion continues for the next few days, probably involving selective diffusion across the NH3-permeable Zellenhaut of the vitelline membrane (Wright and O'Donnell, 2010). Around the time of the second embryonic molt, ammonia production declines abruptly or ceases and the mancae begin to accumulate significant levels of urate. Assuming 6–7 days in the manca stage (Surbida and Wright, 2001), the increase in mean measured urate stores (ca. 4.0 nmoles per manca, or 16 nmoles N) would amply compensate for ammonia excretion given the measured rate in Late Stage 2 (ca. 1.4 nmol d−1). Only very small amounts of urea are excreted (Table 2). There is a modest increase in stored urea from late Stage 2 to mancae (from 0.34 to 0.56 nmol per embryo; Fig. 3), but this represents only about 3% of the nitrogen stored as urate. During the first 4 days following the second embryonic molt, mancae imbibe the marsupial fluid and transition from aqueous gill respiration to aerial respiration (Surbida and Wright, 2001). As they emerge from the marsupium becoming free-living juveniles, they already possess visible pleopodal lungs. The onset of lung respiration probably explains the significant 2.2-fold increase in VCO2 in first instar juveniles relative to mancae (from 3.0 to 6.7 nl min−1), despite their similar masses. Juveniles initiate volatile ammonia excretion following the first 2 days after emergence and rates increase to 0.6–1.2 μmol g−1 h−1 over the next several days (Fig. 5). Rates subsequently decline in second and third instars (2–4 mg in mass) and

3.8. Marsupial fluid volumes Marsupial fluid volumes measured for females carrying different brood stages are given in Table 4. To standardize for female size and number of embryos, values are given as volume per embryo and fractional volume, Vmf / (Ve + Vmf), where Vmf and Ve are the volumes (μl) of marsupial fluid and embryos respectively, assuming embryos to have a density of 1.0 mg μl−1. Fluid volumes declined rapidly in days 1–3 of the manca stage and fluid had essentially disappeared after day 4. Prior to this, results show no significant temporal trend in the marsupial fluid volume per embryo or the fractional volume (single-factor ANOVA: P > .50, df = 8). From these volume data, we estimated that a female with a brood of 50 embryos would have a marsupial fluid volume of approximately 6 μl and used this, together with the measured ammonia excretion rates for the different embryo stages, to estimate the total ammonia levels in the marsupium that would result if excreted ammonia accumulated over time. Values are plotted in Fig. 9 and compared with the measured ammonia concentrations. Marsupial fluid ammonia was highly variable at all stages, ranging from 0 to 24.6 mM, but did not change significantly over time (single-factor ANOVA: F = 0.62, p = .65, df = 47). However, at all stages the mean levels were markedly lower than would be the case assuming no removal of ammonia from the marsupium and using in vitro embryo excretion rates. By L2 and EM, cumulative ammonia excretion assuming retention of ammonia gives a predicted marsupial concentration of 109 mM, 16–24 times higher than the mean measured values. 4. Discussion The results presented here show that early marsupial embryos of Armadillidium vulgare, like adults, are primarily ammonotelic, releasing ammonia into the surrounding marsupial fluid. This is shown by the release of ammonia from embryos reared in vitro in saline, and probably accounts for the significant ammonia concentrations measured in marsupial fluid in vivo (Fig. 9; Surbida and Wright, 2001). Ammonia excretion from embryos in vitro increases to approximately 2 nmol d−1 during Stage 2 as the yolk mass is gradually assimilated into the midgut primordia and somatic tissues. Embryos show very little excretion of 97

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al.

stabilize at 0.1–0.2 nmol g−1 h−1. The elevated rates of ammonia volatilization in early juveniles probably arise from increased protein catabolism. They cannot be explained by catabolism of urate, since the cumulative amount of nitrogen excreted is much too large and wholeanimal urate levels continue to increase, not decrease, during the first and second juvenile instars (Table 3). The same is true for glutamine. Adult terrestrial isopods utilize glutamine as the main intermediary nitrogen store at night, deaminating this to generate volatile ammonia by day (Wieser and Schweizer, 1970; Wright et al., 1994; Wright and Peña-Peralta, 2005). Although we did not explore possible diel variation in ammonia excretion rates in the present study, embryo glutamine concentrations were low (≤ 1 nmol. Embryo−1). Whole-animal glutamine levels increase following marsupial development, from 1.40 ± 0.78 nmol individual−1 in mancae to 6.5 ± 1.9 nmol individual−1 in first-instar juveniles (Fig. 6). Juvenile levels, equivalent to about 9 μmol g−1, are lower than mass-corrected levels measured in adults (26–107 μmol g−1; dusk-dawn comparison, Wright and PeñaPeralta, 2005) but consistent with the intermediary nitrogen storage role seen in adults. Although not examined in this study, other amino acids, particularly glycine, glutamate and arginine, may also play a similar role (Wright et al., 1996). Glutamine concentrations in mancae represent a minor nitrogen reserve relative to other probable intermediary stores. Assuming that urate, urea and glutamine comprise the primary forms of stored nitrogen, their measured levels represent 82%, 5% and 13% of manca nitrogen storage respectively. Mancae of A. vulgare thus primarily exploit uricotelic storage excretion. It remains to be determined whether this is the case in other Oniscidea. The reason for the switch from ammonotely to urate storage excretion in mancae is presently unclear. Marsupial fluid ammonia levels show variable levels throughout development, with values measured here ranging from 0 to 25 mM, and are significantly correlated with maternal hemolymph levels (Surbida and Wright, 2001). 14C inulin injected into the hemolymph appears promptly in the marsupial fluid, confirming high permeability of the maternal sternites and/or cotyledons (Hoese, 1984; Surbida and Wright, 2001). In vitro rates of ammonia production by earlier marsupial embryos would result in very high cumulative ammonia levels if allowed to accumulate (Fig. 9). The fact that large cumulative increases are not observed indicates that embryo ammonia diffuses or is transported into the maternal hemolymph and subsequently excreted. Bromothymol blue staining does not indicate any release of volatile NH3 across the ventral oostegites of gravid females but confirms intermittent volatilization from the pleon, consistent with the previous findings (Wieser et al., 1969; Wright and O'Donnell, 1994). Collectively, the results indicate that maternal metabolism, and not embryo excretion, is the primary determinant of marsupial fluid ammonia levels. The abrupt switch to urate storage excretion in mancae may, however, serve to protect either the brood or the mother from chronically elevated ammonia levels and multiple associated toxic effects (Kaushik et al., 1982; Garcia et al., 2012). In particular, it may protect the mancae from excessive oral intake of ammonia as they consume the marsupial fluid. We observed a further increase in total urate stores in juveniles and urate nitrogen may be recycled later in development as shown in cockroaches (Mullins and Cochran, 1975) and planthoppers (Hongoh and Ishikawa, 1997). In addition, urate may function as an antioxidant, reducing harmful reactive oxygen species (ROS) with the accompanying formation of a urate radical (Becker, 1993; Glantzounis et al., 2005). In this way, urate accumulation could serve to counter possible oxidative damage as mancae transition from their aquatic marsupial habit to air-breathing juveniles, with a sharp increase in metabolic rate (Fig. 8). After exiting the marsupium, A. vulgare shows the predicted allometric scaling of metabolic rate with mass. The observed scaling exponent of 0.739 (95% confidence limits, 0.708 and 0.770), is in close accordance with Kleiber's law (Kleiber, 1932, 1947). The mass-specific VCO2 for a 100 mg A. vulgare at 25 °C, derived from the allometric plot

in Fig. 7, is 5.09 μl min−1 g−1. This shows reasonable accordance with a measured VO2 for this species in normoxic air of 4.78 ± 0.90 μl min−1 g−1 (Wright and Ting, 2006). In summary, marsupial embryos of A. vulgare show significant catabolic metabolism of yolk proteins, accounting for an estimated 7% of total catabolism. Resulting ammonia excretion increases steadily during development until the manca stage when ammonia production falls dramatically and mancae switch to storage excretion of uric acid. This may serve in part to reduce marsupial fluid ammonia levels and protect mancae against toxicity. Urate may also confer antioxidant protection against ROS damage during the 2.2-fold increase in aerobic metabolism as the mancae transition from aquatic to aerial respiration. Juveniles resume ammonotely 2 days after emerging from the marsupium, volatizing the base (NH3) as seen in adults. The experiments herein were carried out in accordance with the National Institutes of Health guide for the care and use of Laboratory animals (NIH Publications No. 8023, revised 1978). Declaration of interest None. Acknowledgments The authors gratefully acknowledge the support of Pomona College and the Howard Hughes Medical Institute through summer fellowships to ERR, KSS, ARG and ZCH. References Akahira, Y., 1956. The function of thoracic processes found in females of the common wood-louse, Porcellio scaber. J. Fac. Sci. Hokkaido Univ. (Ser. VI, Zool.) 12, 493–498. Ban, H., 1950. A culture method of the eggs of a terrestrial isopod, Armadillidium vulgare (Latreille). Sci. Rep. Res. Insts Tohoku Univ. 18 (4), 276–278. Becker, B.F., 1993. Towards the physiological functions of uric acid. Free Radic. Biol. Med. 14, 615–631. Binns, R., 1969. The physiology of the antennal gland of Carcinus maenas (L.). V. Some nitrogenous constituents in the blood and urine. J. Exp. Biol. 51, 41–45. Calvery, H.O., White, A., 1932. Vitellin of hen's egg. J. Biol. Chem. 94, 635–639. Clark, H., Sisken, B., Shannon, J.E., 1957. Excretion of nitrogen by the alligator embryo. J. Cell. Comp. Physiol. 50, 129–134. Fiorini, V.P., 1980. The dorsal organ of arthropods with special reference to Crustacea Malacostraca–a comparative embryological survey. Zoologischer Jarbucher Abteilung fur Anatomie und Ontogenie der Tiere 104 425–265. Fisher, J.R., Eakin, R.E., 1957. Nitrogen excretion in developing chick embryos. J. Embryol. Exp. Morpholog. 5, 215–224. Garcia, L.O., Braun, N., Becker, A.G., Loro, V.L., Baldisserotto, B., 2012. Ammonia excretion at different life stages of silver catfish. Acta Sci. Anim. Sci. 34. https://doi. org/10.4025/actascianimsci.v34i1.11898. Glantzounis, G.K., Tsimoyiannis, E.C., Kappas, A.M., Galaris, D.A., 2005. Uric acid and oxidative stress. Curr. Pharm. Des. 11, 4145–4151. Goodrich, A.L., 1939. The origin and fate of the entoderm elements in the embryogeny of Porcellio laevis Latr. And Armadillidium nasatum B.L. (Isopoda). J. Morphol. 64 (3), 401–429. Greenaway, P., Morris, S., 1989. Adaptations to a terrestrial existence in the robber crab Birgus latro. III. Nitrogenous excretion. J. Exp. Biol. 143, 333–346. Greenway, P., 1991. Nitrogenous excretion in aquatic and terrestrial Crustacea. Mem. Queensland Mus. 31, 215–227. Hartenstein, R., 1968. Nitrogen metabolism in the terrestrial isopod Oniscus asellus. Am. Zool. 8, 507–519. Helden, A.J., Hassall, M., 1998. Phenotypic plasticity in growth and development rates of Armadillidium vulgare (Isopoda: Oniscidea). Israel J. Zool. 44, 379–394. Hoese, B., 1984. The marsupium in terrestrial isopods. In: Sutton, S.L., Holdich, D. (Eds.), The Biology of Terrestrial Isopods. Symposium of the Zoological Society of London. 53. Clarendon Press, Oxford, pp. 65–76. Hoese, B., 1981. Morphologie und Function des Wasserleitugssystems der terrestrichen Isopoden (Crustacea, Isopoda, Oniscoidea). Zoomorphol. 98, 35–67. Hoese, B., Janssen, H.H., 1989. Morphological and physiological studies on the marsupium in terrestrial isopods. Monit. Zool. Ital. Monog. N.S. 4, 153–157. Hongoh, Y., Ishikawa, H., 1997. Uric acid as a nitrogen resource of the brown planthopper Nilaparvata lugens: studies with synthetic diets and aposymbiotic insects. Zool. Sci. 14, 581–586. Hunter, D., Uglow, R., 1993. A technique for the measurement of total ammonia in small volumes of sea-water and haemolymph. Ophelia 37, 31–40. Kashani, G., Waegele, J.-W., Schmalfuss, H., 2011. Redescription of Porcellio brevicaudatus Brandt, 1833 (Isopoda: Oniscidea); with some notes on other synonyms of

98

Comparative Biochemistry and Physiology, Part A 227 (2019) 92–99

E. Rockhill et al. Hemilepistus reaumurii (Milne-Edwards, 1840). Zootaxa 2924, 63–67. Kaushik, S.J., Dabrowski, K., Luquet, P., 1982. Patterns of nitrogen excretion and oxygen consumption during ontogenesis of common carp (Cyprinus carpio). Can. J. Fish. Aquat. Sci. 39, 1095–1105. Kleiber, M., 1932. Body size and metabolism. Hilgardia 6, 315–351. Kleiber, M., 1947. Body size and metabolic rate. Physiol. Rev. 27, 511–541. Linton, S.M., Wright, J.C., Howe, C.G., 2016. Nitrogenous waste metabolism within terrestrial Crustacea, with special reference to purine deposits and their metabolism. In: Weirauch, D., O'Donnell, M.J. (Eds.), Acid-Base Balance and Nitrogen Excretion in Invertebrates. Springer, Cham, Switzerland. Mullins, D.E., Cochran, D.G., 1975. Nitrogen metabolism in the American cockroach. II. An examination of negative nitrogen balance with respect to mobilization of uric acid stores. Comp. Biochemist 50A, 501–510. Munro, A.F., 1953. The ammonia and urea excretion of different species of Amphibia during their development and metamorphosis. Biochem. J. 54, 29–36. Nakamura, M., Wright, J.C., 2013. Discontinuous ammonia excretion and glutamine storage in littoral Oniscidea (Crustacea: Isopoda): testing tidal and circadian models. J. Comp. Physiol. B. 183 (1), 51–59. Okuno, A., Katayama, H., Nagasawa, H., 2000. Partial characterization of vitellin and localization of vitellogenin production in the terrestrial isopod Armadillidium vulgare. Comp. Biochem. Physiol. B 126, 397–407. Packard, G.C., Packard, M.J., 1983. Patterns of nitrogen excretion by embryonic softshell turtles (Trionyx spiniferus) developing in cleidoic eggs. Science 221, 1049–1050. Packard, G.C., Packard, M.J., 1987. Water relations and nitrogen excretion in embryos of the oviparous snake Coluber constrictor. Copeia 1987, 395–406. Rønnestad, I., Thorsen, A., Finn, R.N., 1999. Fish larval nutrition: a review of recent advances in the roles of amino acids. Aquaculture 177, 201–216. Sartori, M.R., Taylor, E.W., Abe, A.S., 2012. Nitrogen excretion during embryonic development of the green iguana, Iguana iguana (Reptilia; Squamata). Comp. Biochem. Physiol. A. 163, 210–214. Schmalfuss, H., 2003. World catalog of terrestrial isopods (Isopoda: Oniscidea). Stuttgarter Beiträge zur Naturkunde, Serie A (Biologie) 654, 1–341. Schmidt, C., 2008. Phylogeny of the terrestrial Isopoda (Oniscidea): a review. Arthropod. Systemat. Phylogen. 66, 191–226. Schmidt-Nielsen, K., 1997. Animal Physiology: Adaptation and Environment, 5th Ed. Cambridge University Press. Shachak, M., Yair, A., 1984. Population dynamics and role of Hemilepistus reaumuri (Audouin and Savigny) in a desert ecosystem. In: Sutton, S.L., Holdich, D.M. (Eds.), The Biology of Terrestrial Isopods. Zoological Society of London Symposia 53. Oxford University Press, pp. 295–312. Sloan, W.C., 1964. The accumulation of nitrogenous compounds in terrestrial and aquatic eggs of prosobranch snails. Biol. Bull. 126, 302–306. Strömberg, J.-O., 1965. On the embryology of the isopod Idotea. Arkiv för Zoologi, Ser. 2 17 (5), 421–473.

Strömberg, J.-O., 1972. Cyathura polita (Crustaea, Isopoda), some embryological notes. Bull. Mar. Sci. 22, 463–482. Surbida, K.L., Wright, J.C., 2001. Embryo tolerance and maternal control of the marsupial environment in Armadillidium vulgare Brandt (Isopoda: Oniscidea). Physiol. Biochem. Zool. 74, 894–906. Warburg, M.R., 1965a. Water relation and internal body temperature of isopods from mesic and xeric habitats. Physiol. Biochem. Zool. 38, 99–109. Warburg, M.R., 1965b. The microclimate in the habitats of two isopod species in southern Arizona. Am. Midl. Nat. 73, 363–375. Weirauch, D., Morris, S., Towle, D.W., 2004. Ammonia excretion in aquatic and terrestrial crabs. J. Exp. Zool. 207, 4491–4504. Wieser, W., Schweizer, G., 1970. A re-examination of the excretion of nitrogen by terrestrial isopods. J. Exp. Biol. 52, 267–274. Wieser, W., Schweizer, G., 1972. Der Gehalt an Ammoniak und Freien Aminosuren sowie die Eigenschaften einer Glutaminase bei Porcellio scaber (Isopoda). J. Comp. Physiol. B. 81, 73–88. Wieser, W., Schweizer, G., Hartenstein, R., 1969. Patterns in the release of gaseous ammonia by terrestrial isopods. Oecologia 3, 390–400. Withers, P.C., 1992. Comparative Animal Physiology. Saunders College Publishing, Fort Worth, TX. Wright, P.A., Fyhn, H.J., 2001. Ontogeny of nitrogen metabolism and excretion. In: Wright, P.A., Anderson, P.M. (Eds.), Nitrogen Excretion. Academic Press, San Diego, pp. 149–200. Wright, J.C., O'Donnell, M.J., 1994. Total ammonia concentration and pH of haemolymph, pleon fluid and maxillary urine in Porcellio scaber Latreille (Isopoda, Oniscidea): relationships to ambient humidity and water vapour uptake. J. Exp. Biol. 176, 233–246. Wright, J.C., O'Donnell, M.J., 2010. In vivo ion fluxes across the eggs of Armadillidium vulgare (Oniscidea: Isopoda): the role of the dorsal organ. Physiol. Biochem. Zool. 83 (4), 587–596. Wright, J.C., Peña-Peralta, M., 2005. Diel variation in ammonia excretion, glutamine levels, and hydration status in 2 species of terrestrial isopod. J. Comp. Physiol. B. 175, 67–75. Wright, J.C., Ting, K., 2006. Respiratory physiology in the Oniscidea: aerobic capacity and the significance of pleopodal lungs. Comp. Biochem. Physiol. A 145, 235–244. Wright, J.C., O'Donnell, M.J., Reichert, J., 1994. Effects of ammonia-loading on Porcellio scaber: glutamine and glutamate excretion, ammonia excretion and toxicity. J. Exp. Biol. 188, 143–157. Wright, J.C., Caveney, S., O'Donnell, M.J., Reichert, J., 1996. Changes in tissue amino acid levels in response to ammonia-stress in the terrestrial isopod Porcellio scaber Latr. J. Exp. Zool. 274, 265–274. Yoshizawa, A., Wright, J.C., 2011. Ionic composition and ion provisioning in marsupial fluid of terrestrial isopods (Isopoda: Oniscidea). Crustaceana 84 (11), 1304–1324.

99