Light, among other factors, allows organisms to synchronise their behaviour with the environment in approximately 24 hours. It does this by influencing the endogenous biological clock, called circadian clock. Circadian rhythms are widely studied due to their impact on the physiology of the organism. Disruption in the circadian rhythm, either by endogenous or external factors can lead to physiological perturbances and diseases. Although mostly studied in higher metazoans, circadian biology is also studied in early branching metazoans including sponges. Sponges, phylum porifera, are known for their remarkable ability to regenerate and remodel their body plan. Moreover, some species have the capacity to reform a fully functional sponge after their cells have been completely dissociated. Herein I explored the influence of light prior and during the cell reaggregation process using the demosponge Amphimedon queenslandica. To achieve this, sponges were exposed either to 12:12 light:dark cycle or constant darkness for one week. The diameter of cell aggregates was used to compute a ratio between big and small aggregates. This ratio was considered to be representative of the aggregation rate. The results obtained reveal no significant difference in the aggregation ratio neither when sponges were exposed to the treatment prior to dissociation, nor when they were exposed during the cell aggregation process.

Light was defined as one of the most dominant factors influencing circadian clocks in the majority of the animals (Beale et al., 2016). The circadian clock is an endogenous molecular mechanism leading to oscillated gene expression in an approximately 24-hour period (Bell-Pedersen et al., 2005). Circadian clocks are found in both unicellular and multicellular organisms, in plants as well as animals, fungi and bacteria (Bell-Pedersen et al., 2005). Even though the circadian rhythm has an endogenous mechanism, environmental cues like temperature, feeding time and light can highly influence the internal clock and lead to its dysregulation when they change (Beale et al., 2016 ; Vetter, 2018).

Although mostly studied in higher metazoans, the circadian rhythm was also found in the basal group represented by sponges. Indeed, circadian genes with diurnal expression have been identified in the demosponge Amphimedon queenslandica and circadian clock proteins have been described in the demosponge Suberites domuncula (Jindrich et al., 2017 ; Muller et al., 2013). Both studies rely on light as an environmental disruptive cue. For instance, the diurnal rhythmic expression of the two genes in A. queenslandica was not detected after sponges were placed for 5 days in constant darkness (Jindrich et al., 2017). Cell reaggregation in S. domuncula was also reported to form larger units when exposed to light in contrast with constant dark treatment (Müller et al., 2012). Understanding how circadian rhythms influence the physiology in basal groups like sponges could further lead to better understand circadian biology in higher metazoans.

Sponges, phylum Porifera, are part of the early branching metazoans. Due to their basal position in the animal tree, sponges are well studied to understanding evolutionary processes and molecular mechanisms that led to multicellularity in higher metazoans (Srivastava et al., 2010). Poriferans display a simple body plan lacking a nervous system and muscles (Brusca et al., 2016). They are mostly marine animals with a biphasic life cycle. Adults are sessile and rely entirely on their aquiferous system for nutrition, excretion and gas exchange (Brusca et al., 2016). One key feature of sponges is their remarkable ability to regenerate and remodel their body plan according to the environment (Lavrov and Kosevich, 2014). Sponges are able to do this thanks to a particular cell type called archaeocytes. Archaeocytes are considered to be sponge stem cells; they are totipotent and have the ability to de-differentiate and redifferentiate into another cell type when needed (Müller and Müller, 1980). Because of the archaeocytes, sponges are able to reproduce asexually either through budding and fragmentation or through gemmules (Brusca et al., 2016). Gemmules are common in freshwater sponges and consist of an aggregate of archaeocytes that is produced when the animal undergoes challenging conditions (Thoms et al., 2008). Once the gemmules are released and the environmental conditions are optimal, archaeocytes are able to differentiate and form a fully functional sponge (Thoms et al., 2008). Similarly, cell aggregates that contain archaeocytes, called primmorphs, are a key element in the formation of a new sponge after cell dissociation.

Cell dissociation and re-aggregation in sponges was reported for the first time in the 19th century by Grant, Bowerbank and Vosmaer (Müller and Müller, 1980) . In the following century, Wilson did an experiment showing the specific reaggregation leading to functional sponges. He cut a sponge into small pieces and squeezed them through fabric to obtain single cells (Wilson, 1907) . He observed the formation of inform cell aggregation which then eventually led to a new sponge. From 1950 on, scientists studied sponges and used cell dissociation to understand mechanisms underlying cell contact and specificity in cell adhesion (Borojevic and Levi, 1964) . Aggregation occurs through single cell movement and aggregation factors which leads to a primary aggregate with no particular structure (Lavrov and Kosevich, 2014 ; Lavrov and Kosevich, 2016). The primary aggregate then undergoes a rearrangement and become a so-called primmorph. Primmorphs then eventually grow and become a new sponge (Lavrov and Kosevich, 2016) . Formation of primary aggregates, development of primmorphs and formation of a new sponge are influenced by many factors including the species, sponge age, temperature, concentration of aggregation factors and light (Lavrov and Kosevich, 2014; Lavrov and Kosevich, 2016 ; Müller et al., 2012). Although light is known as a key factor in the circadian rhythm, its impact on cell aggregation in sponges remains poorly studied.

Herein I explore the impact of light on cell reaggregation in the marine sponge, A. queenslandica. I speculated that sponges exposed to a 12:12 light:dark cycle will have bigger aggregates than sponges exposed to constant darkness. To support this, sponges were either directly dissociated and exposed to two different light treatments for 7 days (long-term experiment) or treated with two different light treatment for 7 days prior cell dissociation (short-term experiment). I quantified the diameter of cell aggregations using ImageJ and computed the ratio of big/small aggregates. My results show no significant difference in the aggregation ratio, neither in the long-term experiment nor in the short-term experiment.

Animals

A. queenslandica specimens were provided by Degnan’s Lab, The University of Queensland. Sponges were collected on Heron Island, southern Great Barrier Reef (-23° 26' 18.71" S, 151° 54' 30.23" E) on the first week of April 2019, during low tide. Sponges were transported to St-Lucia campus, Brisbane and kept in a ~35L aquarium in a recirculation system with a total volume of 3800L. Aquarias were kept at 25°C under an artificial twelve hours light:dark cycle (Challen, C., and Poli, D., personal communication).

Dissociation

Sponge-cell suspension was obtained using a classical mechanical dissociation method (Degnan, B., personal communication). A 4 cc sponge piece was placed in a syringe with a net at the end. Sponge tissue was then pushed through a 250 µm sized mesh into 6mL of sea water in a 60x15 mm Petri dish.

Experimental Design

This project involves 2 different experiments. In total, three sponges were used. Each sponge was cut into four roughly 4 cc pieces in non-sterile conditions using a scalpel and tweezers.

Long-term experiment. Two parts of the parental sponge were dissociated on day 0. Cell dissociation of each part was performed in a 60x15 mm Petri dishe filled with 6mL of sea water. Petri dishes were placed for 7 days either in 12:12 light:dark cycle at 25°C or in the constant dark at 25°C. The diameter of cell aggregates was quantified on day 7 using photographs and ImageJ.

Short-term experiment. Two parts of the parental sponge were placed from day 0 to day 7 in two different ~35L aquaria with either a 12:12 light:dark cycle or constant dark, both at 25°C. On day 7, sponge pieces were dissociated. Cell dissociation of each part was performed in 60x15 mm Petri dishes filled with 6mL of sea water. The diameter of cell aggregates was quantified on day 7, 3.5 hours after cell dissociation using photographs and ImageJ.

The same protocol was repeated for each sponge.

Collecting Data

Each sample was photographed on day 7 using a Olympus SZX9 microscope at a 8x magnificence. Five photographs per sample were taken. Because the diameter of the aggregates was considered to be representative of their size, 20 cell aggregates per photographs were measured using ObjectJ (ImageJ plugin; version 1.52a) to have a total number of 100 diameter measurements per sample. To be quantified, aggregates had to be clearly distinguishable from the background and round-shaped. As there is no recording of aggregates diameter for A. queenslandica allowing me to put a size threshold, the value of 170 µm for the diameter was chosen in the long-term experiment. For the short-term experiment, the threshold was put at a diameter of 110 µm. Those value are based on paper suggesting that primmorphs smaller than 150 µm are not able to develop as well as a paper referencing aggregate diameter at different time point for different species (Lavrov and Kosevich, 2014 ; Eerkes-Medrano et al., 2015).

Aggregates above the threshold were considered as big, and those underneath as small. The ratio of big/small aggregates was calculated to obtain final data.

Statistical Analysis

All statistical analyses and plots were performed using RStudio (Ver 1.0.153).

Long-term experiment. Data for the long-term experiment were considered to be normally distributed because box plots and quantile-quantile diagrams were symmetrical. Since the variance in both treatments was not similar, data were analysed using a two-sided Welch Two Sample t-test.

Short-term experiment. Data for the short-term experiment did not show normal distribution especially in the box plot and quantile-quantile diagrams. Thus, data were analyzed with a non-parametrical two-sided Wilcoxon rank sum test.

Tests were considered to be significant at P < 0.05

Long-term experiment. Sponges were dissociated on day 0 and placed in 12:12 light:dark cycle or constant dark for 7 days (Fig1). The amount of cell per dishes was visually estimated to be equivalent between both treatments. However, the amount of cell between the samples was not equivalent. Indeed, sample 1 (Fig1A. a1,a2) and 3 (Fig1A. c1,c2) have visually a higher amount of cell than sample 2 (Fig1A. b1,b2) at the beginning of the experiment. After one week, 5 samples developed cell aggregates visible without a microscope (Fig1B.a1-a2,c2). Sample 2 in dark condition does not show any aggregates visible with naked eyes (Fig1B. b2).

Most of the cell aggregate have a light color and show distinguishable round shape. The sample 1 in the 12:12 light:dark condition displays a darker colour and less well-defined aggregates (Fig1B. a1). Indeed, the aggregation shows more a fuzzy filamentous cell layer structure than a compact cell clump.

Some samples seem to have a cell decrease. For instance, we can observe a great amount of cell at day 0 in sample 3 (Fig 1A. c1,c2) but there is almost no formation of aggregate one week later for the 12:12 light:dark condition (Fig1B. c1). We can observe the same phenomenon for the sample 2 in the dark condition (Fig1A. b2; Fig1B. b2)

The microscope images are consistent with our primary observation (Fig 2). We can observe small inform aggregates in the 12:12 light:dark condition for sample 1, which reflects the fuzzy structure described previously (Fig2. A1). There are no well-defined cell aggregates. Sample 2 in 12:12 condition show bigger aggregates than sample 1 but as sample 1, they are not well-defined. The shape of the bigger aggregates is fuzzy compared to the small ones in the same photograph (Fig2. B1). Sample 3 for the 12:12 light:dark condition show only small aggregates compared to the same sample in the dark condition (Fig2. C1,C2).

Overall, samples that were placed in the dark tend to show bigger and more well-defined cell aggregates than the samples placed in the 12:12 light:dark cycle (Fig2.A2,C2). Sample 2 for the dark condition, however, displays the smallest and most dispersed aggregates of all samples (Fig2. B2).

Regarding the aggregation diameter distribution, samples that were placed in the light show symmetry in the box plots (Fig3A). There are a few outliers, but the majority of the measurements fit in the box. The sample 1 shows the widest range of diameter going form 19 µm to 575 µm. Sample 2 shows a similar range and is followed by the sample 3, which displays the smallest range, going from 24 µm to 348 µm (Fig3A)

Sponge-cell aggregates in the samples treated with a 12:12 light:dark cycle display smaller diameters than samples treated with constant dark. Samples show more or less the same variation and the boxplots are in the same size range, suggesting that data are distributed normally for the 12:12 light dark condition (Fig 3A, Fig3B).

Samples that were placed in the dark show an asymmetrical distribution of the aggregation diameters, particularly in the sample 1 and 3 (Fig 3B). Outliers are particularly present in sample 1 which also shows the widest range of diameter going form 18 µm to 855 µm. Sample 3 shows the biggest variation in all samples. Overall, samples treated with constant dark tend to show bigger diameters in cell aggregates. However, they show also more variation and boxplots are not in the same size range, suggesting that data for this condition are not distributed normally

After having measured the diameter of a 100 cell aggregates per sample, I performed the ratio of big and small aggregates. Graphs obtained were consistent with the previous observations. Ratios for the 12:12 light:dark condition are similar in all three samples, ranging from 0.05 to 0.07 (Fig4A). Samples 1 and 3 for the dark condition have the highest ratio within all samples which is consistent with the microscope observation (Fig4A). Sample 3 has a ratio of 0.47 and Sample 1 has a ratio of 0.22 . Sample 2 in the dark condition has the lowest value within all sample with a ratio of 0.01 (Fig4A).

The mean ratio of the samples treated with constant darkness is almost 5 fold greater than the mean of the samples treated with a 12:12 light:dark cycle (Fig4B). However, this difference was not considered to be statistically significant (P=0.13).

Short term experiment. Sponge pieces were kept entire form day 0 to day 7 and were placed either in an aquarium with 12:12 light:dark cycle or constant dark. On day 7, sponge pieces were dissociated and aggregates quantified.

All samples display a grey turbid solution directly after cell dissociation (Fig 5A). Cell density was visually estimated to be equivalent. However, one sample seems to be less concentrated than the others (Fig5A. b1).

After 3.5 hours, sponge-cell aggregates were visible by naked eyes (Fig5 B). Indeed, samples that were treated with 12:12 light:dark cycle prior dissociation show a stronger pattern than the ones treated with dark (Fig5B), especially the first sample (Fig5B. a1).

Microscopy images reflect our visual observations. In fact, samples that were treated with a 12:12 light:dark cycle show bigger and rounder primary cell aggregates (Fig6. A1-C1). Samples that were kept in the dark for one week also show aggregates, but the size is smaller and the shape not as round. (Fig6 A2-C2).

The diameter distribution between the sample that were placed in the 12:12 light:dark cycle and the ones place in constant dark prior dissociation shows less variance than in the long-term experiment (Fig7) Indeed, all samples that were placed in the light show symmetry in the box plots and very few outliers (Fig7A). The sample 3 shows the widest range of diameters going form 46 µm to 785 µm. Sample 1 shows a range going from 50µm to 364 µm followed by sample 2, which displays the smallest range, going from 15 µm to 177 µm (Fig7A)

Sponge cell aggregates diameter in samples treated with a 12:12 light:dark cycle display bigger diameters than samples treated with constant dark. In addition, they show more or less the same variation and the box plots are in the same size range, suggesting that data are distributed normally for this condition (Fig 7A,).

Samples that were placed in the dark prior dissociation also show symmetry in the boxplots and few outliers (Fig7B). Diameter rang goes form 33 µm to 278 µm for sample 3 which displays the widest range. Sample 1 and 2 show almost the same range going form 39 µm to 168 µm and 43µm to 171 µm respectively. The variation of the samples seems to be more or less the same. In addition, the box size is also more or less the same, suggesting that data for this condition are also normally distributed.

After having measured the cell aggregate diameters, I compute the ratio for all sample. Overall, the aggregate ratio was greater in the samples treated with 12:12 light:dark cycle prior dissociation than in samples treated with constant dark. (Fig8A). Indeed, The first sample treated with 12:12 light:dark cycle has a ratio of 0.62, which is the highest in all sample. Sample 2 and 3 show a similar ratio regarding the 12:12 light:dark treatment with 0.29 and 0.32 (Fig8A). Samples treated with constant dark prior dissociation show a similar ratio going from 0.14 for the first sample to 0.17 for the second sample and 0.16 for the third one.

In contrast to the long-term experiment, the mean ratio of the samples treated with 12:12 light:dark cycle for 7 days prior dissociation is greater than the mean ratio of samples treated with constant dark (Fig8B). 12:12 light:dark cycle condition has a mean ratio of 0.41 and the dark condition a mean ratio of 0.16. Nonetheless, this difference is not considered to be statistically significant (P= 0.081).

Sponges have been shown to be able to detect light in the larval stage as well as in the adult stage (Muller et al., 2013; Rivera et al., 2012). It has also been shown that light has an impact on those organisms, influencing their circadian physiology (Muller et al., 2013; Jindrich et al., 2017).

Herein I investigated the impact of light and dark prior and during the cell re-aggregation process in the demosponge A. queenslandica.

Overall in both experiments, the second sponge is the one that shows the lowest ratio. The fact that it is the same sponge in the long-term and short-term experiment suggest that the sponge behaves differently from the first and third sponge. This difference could be due to the sponge's age or condition. Factors such as the age of the organism or the captivity time have a negative influence on cell reaggregation (Lavrov and Kosevich, 2014) and should be taken into account.

The results obtained for the long-term experiment do not support my hypothesis. Indeed, I expected to see a greater amount of big cell aggregates in the 12:12 light:dark condition than in the constant dark. Even though no statistical difference was found between both treatments, my results tend to show the exact opposite; bigger aggregates were present in the samples exposed to dark.

The results could be explained by the fact that at the beginning of this experiment, cells were not in a cell-suspension. Indeed, due to technical issues, cells had to be centrifuged to be concentrated and were then place in Petri dishes. This additional step could have influenced the viability of the cell and thus their ability to reaggregate with each other.

Another aspect that probably influenced the results was the presence of ciliates in the samples. Indeed, ciliates feed on sponge cells and may thus influence the cell concentration as well as the aggregates size (Degnan, B., personal communication). For instance, a visual cell decrease was observed in 12:12 light:dark treatment for the sample 3 in which ciliates were particularly dominant. Although also present in the sample exposed to darkness, ciliates do not seem to have developed as well as in the light:dark treatments, thus affecting less the aggregation ratio. Their presence may also have impacted the initial cell concentration which is a factor influencing aggregate size. In fact, the higher the cell concentration at the beginning, the bigger the cell aggregates (Sipkema et al., 2003).The initial low sponge-cell concentration as well as the presence of the ciliates may explain the fact that almost no aggregation was observed in the second sample.

For future experiments, it would be useful to perform staining with DAPI and trypan blue to distinguish living from dead cells (Chan et al., 2017).

Regarding the short-term experiment, the results neither support my hypothesis nor the literature. Indeed, a study by Müller et al., 2012, showed that three-dimensionally growing cell aggregates exposed to darkness reaggregate slower and had less well-defined primmorphs morphology than cell aggregates treated with light. This observation was linked with a decrease in glycogen. Glycogen synthesis is initiated by glycogenin, which is repressed by the circadian clock protein nocturine. For future experiments, it would be interesting to look at the glycogene and nocturine levels in A. queenlandica primmorphs to assess whether this observation is also true in another sponge species. It would also be interesting to target other important metabolic enzymes to look which ones are under circadian control or not.

One aspect that could be investigated is whether the production of aggregation factors is influenced by the circadian clock. Aggregation factors are macromolecules that are released when a sponge is dissociated and allows binding between cells (Lavrov and Kosevich, 2014). The fact that cells reaggregate in a slower manner in dark condition could maybe also be due to a lack of production of aggregation factor.

As this experiment is a first trial, results should be interpreted carefully. For instance, the thresholds chosen for the experiments rely on approximations of primmorphs size of other experiments using different species (Rivera et al., 2012). Before conducting other experiments on primmorphs size in A. queenslandica, it would be useful to quantify and report aggregates size over time in this species. This would allow to set a precise threshold and thus archiving more accurate comparison in further experiments.

The lack of statistical difference between both light treatments could also be explained by the ability of the circadian clock to maintain the rhythmic activity for a while even if the external factors are removed (Bell-Pedersen et al., 2005) . In A. queenslandica for example, the detection of rhythmicity in certain gene expression takes at least five days to dissipate when the animal is exposed to constant dark (Jindrich et al., 2017). On that basis, I can suspect that cells behaved the same way for at least five days in the long-term experiment before being affected by the treatments. It would be interesting to conduct the same experiment with mesurment over time and for a longer period to assess whether there is an aggregation size plateau reached by the cells exposed to constant dark after a few days. Moreover, it would be useful to perform a PCR on circadian genes to see if their expression or lack of expression could be correlated with a change in primmorphs morphology. This could also be useful to assess whether the sponges are strictly dependent on the environmental cues for their rhythmicity or not.

Over all, the influence of light on cell reagreggation should be further investigated. Even though the aggregation leading to a fully functional sponge is not found throughout all species (Eerkes-Medrano et al., 2015), understanding its process is important. As the sponges culture form aggregates remains difficult (Muller et al., 2004), knowing which factors can promote or disrupt the process could help to improve the culture protocol .This would be of great value in the medical field where sponges are known to produce lots of chemical fighting virus, bacteria and cancer (Anjum et al., 2016).

I would like to particularly acknowledge Professors Sandie and Bernard Degnan for providing me the study specimens, for their help and good advice throughout the experiment. None of it would have been possible without their support.

I would like to thank Eunice Wong and Davide Poli (Prac Tutors) for their help in the execution of the experiment.

Finally, I would like to thank Nicola Blackmore for the proofreading of the manuscript.

Anjum, K., Abbas, S.Q., Shah, S.A.A., Akhter, N., Batool, S., and Hassan, S.S. ul (2016). Marine Sponges as a Drug Treasure. Biomolecules & Therapeutics 24, 347–362.

Beale, A.D., Whitmore, D., and Moran, D. (2016). Life in a dark biosphere: a review of circadian physiology in “arrhythmic” environments. J Comp Physiol B 186, 947–968.

Bell-Pedersen, D., Cassone, V.M., Earnest, D.J., Golden, S.S., Hardin, P.E., Thomas, T.L., and Zoran, M.J. (2005). Circadian rhythms from multiple oscillators: lessons from diverse organisms. Nat Rev Genet 6, 544–556.

Borojevic, R., and Levi, C. (1964). Etude au microscope electronique des cellules de l’eponge:Ophlitaspongia seriata (Grant), au cours de la réorganisation après dissociation. Z.Zellforsch 64, 708–725.

Brusca, R.C., Moore, W., and Shuster, S.M. (2016). Invertebrates (Sunderland, Massachusetts U.S.A: Sinauer Associates, Inc., Publishers).

Chan, L.L.-Y., McCulley, K.J., and Kessel, S.L. (2017). Assessment of Cell Viability with Single-, Dual-, and Multi-Staining Methods Using Image Cytometry. In Cell Viability Assays, D.F. Gilbert, and O. Friedrich, eds. (New York, NY: Springer New York), pp. 27–41.

Eerkes-Medrano, D., Feehan, C.J., and Leys, S.P. (2015). Sponge cell aggregation: checkpoints in development indicate a high level of organismal complexity. Invertebr Biol 134, 1–18.

Jindrich, K., Roper, K.E., Lemon, S., Degnan, B.M., Reitzel, A.M., and Degnan, S.M. (2017). Origin of the Animal Circadian Clock: Diurnal and Light-Entrained Gene Expression in the Sponge Amphimedon queenslandica. Front. Mar. Sci. 4, 327.

Lavrov, A.I., and Kosevich, I.A. (2014). Sponge cell reaggregation: Mechanisms and dynamics of the process. Russ J Dev Biol 45, 205–223.

Lavrov, A.I., and Kosevich, I.A. (2016). Sponge cell reaggregation: Cellular structure and morphogenetic potencies of multicellular aggregates: sponge cell reaggregation: cellular structure. J. Exp. Zool. 325, 158–177.

Müller, E.G., and Müller, I. (1980). Sponge cell aggregation. Mol. Cell. Biochem. 29, 131–143.

Müller, W.E.G., Wang, X., Grebenjuk, V.A., Korzhev, M., Wiens, M., Schlossmacher, U., and Schröder, H.C. (2012). Nocturnin in the demosponge Suberites domuncula : a potential circadian clock protein controlling glycogenin synthesis in sponges. Biochem. J. 448, 233–242.

Muller, W.E.G., Schroder, H.C., Pisignano, D., Markl, J.S., and Wang, X. (2013). Metazoan Circadian Rhythm: Toward an Understanding of a Light-Based Zeitgeber in Sponges. Integrative and Comparative Biology 53, 103–117.

Müller, WernerE.G., Grebenjuk, VladislavA., Le Pennec, G., Schröder, H.-C., Brümmer, F., Hentschel, U., Müller, IsabelM., and Breter, H.-J. (2004). Sustainable Production of Bioactive Compounds by Sponges?Cell Culture and Gene Cluster Approach: A Review. Mar. Biotechnol. 6.

Rivera, A.S., Ozturk, N., Fahey, B., Plachetzki, D.C., Degnan, B.M., Sancar, A., and Oakley, T.H. (2012). Blue-light-receptive cryptochrome is expressed in a sponge eye lacking neurons and opsin. J. Exp. Biol. 215, 1278–1286.

Sipkema, D., van Wielink, R., van Lammeren, A. a. M., Tramper, J., Osinga, R., and Wijffels, R.H. (2003). Primmorphs from seven marine sponges: formation and structure. J. Biotechnol. 100, 127–139.

Srivastava, M., Simakov, O., Chapman, J., Fahey, B., Gauthier, M.E.A., Mitros, T., Richards, G.S., Conaco, C., Dacre, M., Hellsten, U., et al. (2010). The Amphimedon queenslandica genome and the evolution of animal complexity. Nature 466, 720–726.

Thoms, C., Hentschel, U., Schmitt, S., and Schupp, P.J. (2008). Rapid tissue reduction and recovery in the sponge Aplysinella sp. Mar Biol 156, 141–153.

Vetter, C. (2018). Circadian disruption: What do we actually mean? Eur J Neurosci.

Wilson, H.V. (1907). On some phenomena of coalescence and regeneration in sponges. J. Exp. Zool. 5, 245–258.