Growth, filtration and respiration characteristics of small single-osculum demosponge Halichondria panicea explants Open Access

All demosponges, except a few carnivorous ones (Vacelet and Boury-Esnault, 1995; Lee et al., 2012), are filter-feeding modular organisms that consist of a set of repetitive aquiferous units or modules with one water-exit osculum per module (Fry, 1970; Ereskovskii, 2002; Kealy et al., 2019; Kumala et al., 2023). However, some large tropical single-osculum sponges are composed of many modules, each with a true osculum that opens into a common spongocoel, which opens to the ambient water via a static pseudo-osculum (Riisgård and Larsen, 2022b). Once a mature module has been formed, its filtration (F) and respiration (R) rates no longer change over time. Therefore, when a sponge grows to become a large ‘population of modules’, both the total filtration and respiration rates of the sponge increase linearly with the increasing number of mature modules, i.e. F=a₁W^b₁ and R=a₂W^b₂ where b₁=b₂=1. According to the bioenergetics growth model presented by Riisgård and Larsen (2022a), the weight-specific growth rate µ=(1/W)dW/dt=aW^b₁/W, which for b₁=1 becomes µ=a, which is a constant and the growth is therefore exponential according to definition.

Sponge modules only grow to a certain size because of increasing resistance to flow in the increasingly longer inhalant and exhalant canals. Further, an increasing relative content of water compared with sponge-body tissue puts an upper limit on the size of the module for it to remain a stable structure. Thus, for a sponge to increase in size, new modules must be formed, which leads to a multi-oscular sponge that consists of single-oscular modules that have reached their maximal size and hence their maximal filtration rates. This implies the scaling F/V is constant, because the total sponge volume equals V=number of modules×volume of module (see fig. 3 of Kumala et al., 2023). This also implies that both filtration and respiration rates increase linearly with V in multi-oscular sponges. However, the growth characteristics of a small single-osculum module sponge are fundamentally different.

Reiswig (1975) studied the aquiferous system elements in the marine demosponges Halichondria panicea and Haliclona permollis and noticed variations in tissue densities and canal systems near ‘growth points’ (tips of tubular units or branches) compared with ‘mature’ regions with relatively stable dimensions. We believe that small growing single-osculum single-module explants of H. panicea are comparable to ‘growth points’ in the large multi-oscula multi-modular sponge. In this way by using single-osculum explants, we can study the growth, filtration and respiration characteristics of sponge ‘growth points’, which is of fundamental importance for understanding the growth and bioenergetics of sponges.

A scaling relationship between filtration rate (F) and body volume (V) of a single-osculum module was derived by Riisgård and Larsen (2022b) by considering a single inhalant canal of length L whose wall with water-pumping choanocyte chambers (CCs) separates it from the exhalant canal system. The filtration rate created by the CCs is determined by the product of the filtration rate (F_CC) of each CC and their number, which is proportional to the wall area, i.e. F∝L², while the volume scales as V∝L³ from which it follows that F∝(V^1/3)² or F=aV^2/3. Hence, the volume-specific filtration rate scales as F/V∝V^2/3–1=V^–1/3 (Riisgård and Larsen, 2022b), indicating a decrease with increasing V when a small single-osculum sponge module grows bigger. The predicted scaling was supported by data from the literature of experimentally measured F and V in single-osculum H. panicea explants (see fig. 2 of Riisgård and Larsen, 2022b). Here, we further studied filtration and respiration rates, volume-specific and weight-specific filtration rates, F/R ratios and body growth rates of small single-osculum H. panicea demosponge explants of various sizes exposed to different concentrations of algal cells.

It has recently been pointed out by Riisgård and Larsen (2022a) that in filter-feeding marine invertebrates, the F/R ratio is generally high (i.e. >10 l of water filtered per ml oxygen consumed) in filter-feeding marine invertebrates and that the oxygen extraction efficiency is consequently low, typically ≤1% (Jørgensen et al., 1986). Riisgård and Larsen (2022a) suggested that when the biomass of a growing filter feeder increases, the total respiration consequently increases, but the increasing oxygen demand is easily met by an increase in oxygen extraction efficiency. However, a growing animal must also increase its filtration rate, and thus the ingestion of food to cover its growth and respiratory needs. This ensures that the F/R ratio remains unchanged because a reduction in the F/R ratio would cause starvation. Therefore, Riisgård and Larsen (2022a) suggested – as a theory – that the b-exponents in the equations for filtration and respiration, F=a₁W^b₁ and R=a₂W^b₂, may (during evolution) have become near equal depending on species and adaptation to living site. Because of the simple structure of sponges, which have no organs or real tissue that may store energy reserves to overcome starvation periods, the exponents for F and R versus W tend to be close to 1 as seen in multi-oscular multi-module sponges (Reiswig, 1974; Thomassen and Riisgård, 1995; Southwell et al., 2008; Fiore et al., 2013; McMurray et al., 2014; Ludeman et al., 2017; Goldstein et al., 2019; Dahihande and Thakur, 2019; Kumala et al., 2023; Lesser, 2023). However, little is known about b-exponents in small growing single-osculum sponges, and therefore one objective of the present study was to test the above hypothesis of b₁≈b₂≈1 by means of experimental data.

It is well known that sponges may undergo periodic cessation of filtering activity (Reiswig, 1971) or reduce filtration activity as a result of disturbance (Riisgård et al., 2016), but the b-exponents presented in the present study are based on optimally filtering sponges.

Because in situ (wet) sponge modules of volume V have a density essentially equal to that of seawater, the volume V∝W_wet (wet weight) and their dry weight (W) were calculated by the power law relationship W(W_wet). Conversion of algal cells to chlorophyll a concentration was calculated as 1 µg Chl a l⁻¹=799 R. salina cells ml⁻¹ (Clausen and Riisgård, 1996).

Table 1 shows a summary of the key parameters of the growth experiments. Fig. 1 is an example of the growth of a sponge module fed with a certain algal concentration for 38 days and photographed weekly to determine the parameters needed for estimating module volume (V_mod) according to Eqn 3. The growth history of the total sponge module volume (V) under different food treatments is shown in Fig. 2 and the corresponding regression equations are listed in Table 1. It can be seen that the growth rate increases with increasing algal concentration, with a maximum mean volume-specific growth rate of 4.2% day⁻¹. The relationship between volume-specific filtration rate and total volume of sponge modules is shown in Fig. 3, where the power function relationship F/V=7.08V^–0.24 is not far from the theoretical scaling relationship F/V∝V^−1/3 described in the Introduction, although the power curve fit (R²=0.211) to the scattered data is less successful than a simple but theoretically unfounded linear regression line (R²=0.327). The relationship between the dry and wet weights of sponge modules is shown in Fig. 4. The power function W=0.047W_wet^0.73 indicates that an increasing part of the module volume is made up of water as the module grows bigger, which was used in deriving the scaling relationship for F(V). Further, using the power function for W, measured filtration and respiration data can be expressed in the usual way in terms of dry weight rather than volume. As shown in Fig. 5, F=a₁W^b₁, b₁=1.047 and R=a₂W^b₂, b₂=0.682, which suggests that b₁≈1 (in agreement with the working hypothesis) and that b₂<1. The data in Fig. 6 were obtained by applying Eqns 4 and 5 to dry weight data taken from converted total volumes in Table 2. The short time records of only three dates led to only two points at each food treatment in the plot of weight-specific growth rate µ=(1/W)dW/dt=aW^b shown in Fig. S2, hence a considerable uncertainty, yet the expected trend of increasing µ with increasing Chl a concentration (C) is clear. A plot of µ versus C (Fig. 6) suggests a trend towards an upper maximum at high values of C.

The F/R ratio versus dry weight (W) is shown in Fig. 7A and the F/W ratio versus W in Fig. 7B. The very low R² value of 0.067 in Fig. 7A indicates no relationship, i.e. the F/R ratio is constant and independent of sponge size in agreement with the hypothesis for filter-feeding marine invertebrates that b₁≈b₂, and further, in the present case of a small growing sponge modules, that b₁≈b₂≈1. Thisis also confirmed by the lack of a relationship between F/W and W (Fig. 7B).

In the present study, the mean F/R ratio was found to be 7.0±3.1 l of water filtered per millilitre of oxygen respired, which may be compared to a value of 11 reported by Riisgård et al. (1993), 2.7 reported by Thomassen and Riisgård (1995) and 15.5 l H₂O ml⁻¹ O₂ reported by Riisgård et al. (2016). This last value indicates that the F/R ratio of mature H. panicea is generally above the minimum reference value of 10 l H₂O ml⁻¹ O₂, but also that the ratio may be influenced by the experimental conditions, e.g. disturbance, water quality and food concentration. Thus, spontaneous contractions causing variations in filtration activity of single-osculum explants of H. panicea have been documented by, for example, Goldstein et al. (2019) and Kumala et al. (2023). In the present study, where the mean filtration rate (F) was measured on up to 18 modules, spontaneous contractions among modules probably reduced the mean filtration rate, whereas the mean total respiration rate (R) remained more stable, resulting in some scatter in the F/R data shown in Fig. 7A.

The scaling relationship F/V∝V^–1/3 of a single-osculum module derived by Riisgård and Larsen (2022b) was based on the suggestion that F is determined by the product of the filtration rate of each choanocyte chamber and their number, which is proportional to the wall area between the inhalant and exhalant canals. It appears from Fig. S2 that the exponent b in the growth relationship µ=aW^b is close to zero, which according to the bioenergetic growth model (see Eqn 3 of Riisgård et al., 2014) indicates that µ≈constant, and therefore b₁≈b₂≈1, which is exponential growth.

It can be concluded that the present data support the theoretical relationship F/V∝V^–1/3 and that b₁≈b₂≈1 in small growing single-osculum explants.

Thanks are due to Josephine Goldstein for skilful technical assistance and help with data analysis. We acknowledge the comments made by the editor and an anonymous reviewer that improved an earlier version of the manuscript. All experimental work was carried out using facilities at the University of Southern Denmark.