The Egg Packing Pressure Index of Calanoid Copepod as a Novel Eco-Indicator in Diverse Geographical Ecosystems

Highlights

What is the main finding of this study?

• It is the first confirmation of a true egg sac membrane through a multigenerational assay.
• Packing pressure index depends on geographical locations rather than genetic similarities.
• Packing constraints increase with increasing temperature and salinity.

What is the main implication of this finding?

• An index is introduced to gauge the reproductive effort and maximal reproductive capacity of copepods.
• Packing constraints provide an as yet unrecognized indicator for ecosystem health.

Abstract

Egg-carrying ectothermic planktonic organisms in stressed conditions adapt diverse reproductive trade-off strategies, which are denoted by markers of stress, either in their physiology, morphology or reproductive characteristics. This is the first study documenting the fact that egg packing constraints can provide a novel marker of stress through experimental and field data, attuned by a remarkable bioindicator egg-bearing copepod species, Eurytemora affinis, in diverse physical conditions and transboundary sites through a multigenerational assay. This study propounds a packing pressure index (PP index), which is an efficient tool for demarcating reproductive efforts along with reference lines of packing constraints in bioindicator organisms. The packing pressure index for E. affinis varied across geographical locations, showing distinct north–south variations, along with a strong packing constraint in higher-temperature and -salinity conditions. The novel relationships between morphological and reproductive traits and packing constraints could be standardized. Ensuring the presence of the egg sac membrane and its relationship with varied physicochemical parameters can aid in developing a comprehensive understanding of reproductive strategies in keystone species like E. affinis.

1. Introduction

Calanoid copepods are divided into two groups according to their reproductive modes: broadcast spawners or free-spawning copepods, and egg-breeding or -bearing copepods. In the latter, the female copepod retains her eggs until they hatch in a sac attached to the genital somite [1]. The changes in the state of the ocean, such as tectonic movements, oxygen profiles and the primary and secondary productivity of lower trophic organisms, contributed significantly to the evolution and ecological adaptation of calanoid copepods [2]. In addition, ecological trade-offs in reproductive and life history traits since the Paleozoic era paved an evolutionary track, particularly for the egg-bearing and free-spawning copepods [3]. An ecological effort, often termed parental investment, was noticed in egg-bearing species, corroborating a higher rate of offspring survival and protection [4]. The shift from egg-bearing copepod species to free-spawners occurred due to a range of events, which have their respective advantages and disadvantages [5]. The prime example of an evolutionary event is the transitional change in reproductive strategies depending upon several environmental parameters, such as the availability of nutrients, turbulence of the ocean, pressure from predators and trade-offs in life history traits [6,7]. Nonetheless, although the effort required from egg-bearing copepod species is higher in terms of maintaining the clutch than the effort required for free-spawning species, they show an ability to secure higher survival rates in subsequent generations and also an extended life span for the parental generation [8,9]. Free spawners, on the other hand, show a higher growth rate and a lower life span, with a large number of eggs produced but a higher predation risk. The evolution of egg-bearing and free-spawning copepods predominantly reflects an adaptation to their surroundings, where egg-bearing copepods are well adapted to the high predation risk and lower resource availability; free-spawners are attuned to turbulent and nutrient-rich environments [10].

Previous studies on Calanoida regarding their clutch size and egg size often overlooked any surveillance of the egg sac or egg packing associated with the morphology and reproductive strategies of the female population [11,12,13,14]. The egg sac has a very important role in the life history of egg-bearing copepods since it contributes to the survival of the species by protecting the carrying eggs [3] and, on the contrary, also contributes to the mortality of the species by increasing the risk of predation. In fact, ovigerous females are more exposed to visual predators [15] and have a lower escape capability, which leads to a higher predation risk [3,16]. In addition, Logerwell and Ohman [3] showed the importance of the egg sac position and found that the interaction with swimming behavior influences the predation risk more than the body size.

Some authors doubted the presence of a true egg sac with an external membrane in calanoid copepods and stated that it was an egg mass formed by secretions that paste the eggs together and attach the mass to the genital somite [17,18]. On the contrary, other authors showed that there is a real egg sac in the calanoid copepod, with clear evidence of the presence of an external membrane, which induces the behavior of packing the eggs in the sac [1,19,20]. Similar to many taxa where females retain their eggs in a body cavity (e.g., Cladocera), the egg-bearing copepods face the constraints of egg packing, which can affect their egg size, clutch size and also the reproductive trade-off [21,22,23]. This reproductive trade-off is a crucial phenomenon, which provides a theoretical understanding of the pattern of resource allocation in the life history of an organism. This can be influenced by the combination of all the data of the prosome and clutch volume in stressed conditions due to altered environmental factors, such as temperature, salinity, light, pollutants, etc. Reproductive effort in an organism is a manner that marks the energy investments and their trade-off for performing diverse physiological functions. Beck and Beck (2005) [23] proposed the first study on the effect of reproductive allocation in relation to the egg size for turtles and copepods, reporting that the nature of this relationship depends on the presence of packing constraint. For copepods, Beck and Beck (2005) [23] analyzed different taxonomic groups, including divergent branches, such as parasitic and free-living copepods, by using composite phylogenies. However, mixing different phylogenetic levels in Copepoda might not be an appropriate tool to analyze the role of the packing constraints and, hence, we need to focus on bioindicator organisms from lower taxonomic levels. We hypothesize that the packing constraints should concern primarily egg-bearing copepods because broadcast spawners do not carry their eggs.

Among the brooding species, calanoid copepod Eurytemora affinis, a sentinel species [24] and a remarkable bioindicator to any stressor, offers a good model to test the presence of egg packing constraints in ovigerous females given its wide ecological significance [14,25]. This egg-bearing copepod is a dominant species in estuaries and salt marshes of the Northern hemisphere, from the coastal regions of North America to Europe and Asia [26]. Lee and Frost (2002) [27] stated that E. affinis is a species complex composed of distinct populations, which comprises divergent clades between North America and Europe at the gene [28,29] and physiological response levels [26,30]. The divergence among this cryptic species complex is due to an evolutionary separation, inducing an adaptation to their highly fluctuating habitats [14]. In a previous study, Souissi et al. (2021) [31] used experimental data and confirmed the presence of a trade-off between egg number and egg size in E. affinis. The former study suggested that the decision of reproductive energy allocation in this species could be simultaneous and not sequential, and that the observed reproductive trade-off could disappear under very stressful conditions, such as high temperature. Furthermore, the secretion of the egg sac membrane to protect all eggs should match the reproductive effort (or clutch volume). Consequently, the presence of any packing constraint in E. affinis could be an indication of an evolutionary reproductive strategy, allowing for the possibility to adjust the reproductive effort, even during the last stages of reproduction. In response to constraints such as temperature, E. affinis tends to adjust the number of eggs it produces per clutch or alter the size of the eggs, favoring fewer but larger eggs that have a higher chance of surviving in a stressed environment. This is an example of an adaptive strategy where the copepod might invest in fewer, but more robust, offspring that have a higher probability of survival in a deteriorating environment. On the contrary, trends were also shown that copepods might produce a higher number of eggs, to ensure a maximum number of hatchlings being released to secure the population density for the next generation, depending on various allied factors, such as water quality parameters and the availability of food sources, through an adjustment of egg size and number [14,31,32]. In addition, if environmental conditions begin to change (e.g., temperature, food availability, or the presence of predators), E. affinis could delay or speed up egg production based on the likelihood of favorable conditions for offspring survival. This is a form of reproductive timing adjustment, allowing the copepod to maximize its reproductive output when conditions are ideal and reduce efforts when conditions are less favorable. In order to test the utility of using packing constraint as a marker of temperature stress, we combined a significant experimental database obtained from controlled conditions (Souissi et al., 2021) [31] with another database from different field conditions. In this study, our aim is to report the presence of egg packing constraints enclosed in an external membrane in relation to the strategy of the ovigerous female of E. affinis. The primary objective is to comprehend the relationship between the true egg sac and the genital somite with the effects of temperature on egg packing constraints. The genital somite accommodates the reproductive organs, including the gonads and oviducts. A wider genital somite can provide more space for larger or more eggs, allowing the copepod to invest more energy in reproduction. This could lead to an increase in the number of eggs produced (fecundity) or the size of each egg. This can affect the copepod’s reproductive success and survival as well as fitness rates of the offspring [33]. In addition, our tangential aim is to comprehend the interpopulation and/or inter-clade differences in egg packing constraints of E. affinis.

2. Material and Methods

2.1. Eurytemora affinis: Isolation and Culture

Specimens of the cryptic species complex E. affinis were collected from different estuaries in Europe and North America (

Table 1. Sampling sites of the cryptic species complex E. affinis with the latitudes, the dates, the number of observations and the sources of the samples.

Continent	Estuary		Latitude–Longitude	Dates (No. of Observations)
				2006	2007	2008	2009	Other Years
Europe	Elbe		53°32′06 N–09°47′31 E	17 March (20)
	Scheldt		51°21′06 N–04°14′58 E	4 April (19)	11 April (11)	6 April (11)
	Seine		49°28′33 N–00°27′54 W	23 May (19)	10 October (21)–6 November (16)	8 March (19)–10 April (27) 8 June (34)–8 July (21) 20 September (11)–4 November (15)	25 February (20)	10 February 2010 (18)
	Loire		47°17′23 N–02°01′52 W	18 April (20)	15 November (31)	16 April (9)	25 June (16)
	Gironde		45°14′80 N–00°43′50 W	20 April (20)			15 April (15)
North America	St La w rence	Isle Verte	48°00′20 N–69°25′50 W		18 May (15)–12 June(20) 11 July (26)		22 May (7)
		St Jean Port Joli	47°12′59 N–70°16′22 W		11 July (6)	15 July (19)	20 May (20) 2 June (33)
		Montmagny	46°59′26 N–70°33′13 W		14 June (20)		May (7)
		Berthier sur Mer	46°56′07 N–70°44′07 W		11 July (1)	15 July (17)	29 May (5)
	Chesapeake Bay		39°23.81′ N–76°03.32′ W			16 April (10)–15 December (19)	6 February (22) –28 April (11)	May 2002 (6) April 2003 (4)

) and are representative of the northern transatlantic distribution of the species in estuaries (Figure 1, Table 1). In field conditions, sampling was conducted using plankton nets with different sizes (e.g., WP2 plankton net used in the Seine estuary) depending on the depths of the sampling sites. Thereafter, zooplankton samples were gently cleaned with the water sampled from the same site to remove excess suspended matter, if any. The samples were fixed by using diluted ethanol until analysis. In the lab, after rigorous homogenization, a sub-sample was used to sort the required number of ovigerous females. The ovigerous females were then gently cleaned and put in synthetic water, with the salinity corresponding to the field conditions and preserved with diluted ethanol in Eppendorf tubes. The highest number of specimens came from the Seine estuary, which we consider in this study the reference site for the E. affinis population because of its position, lying at an intermediate latitude in Europe (Figure 1). Apart from being situated at an intermediate latitude of Europe, the Seine estuary is a pivotal location in terms of its ecology, pollution baselines, and the dynamics of environmental parameters [14,25,34]. The Seine estuary has a high biodiversity and species richness of organisms, with E. affinis showing the highest abundance of all copepod species. In addition, it marks one of the busiest and most industrious shipping routes, with heavy commercial pursuits and anthropogenic inputs from nearby megacities [34]. This area is also characterized by high hydrodynamic forces due to its topography and coastal constructions for human needs [34]. In addition to the in situ samples, strains of E. affinis population originating from the Seine estuary were also cultured in different conditions of temperature and salinity. Studying copepods from lab cultivation and their natural habitat allowed us to infer the variability between cultivated and in situ copepods, additionally making this study more ecological and environmentally relevant. Furthermore, this comparison shed light on diverse geographical ecosystems, such as the estuaries in Europe and North America. In fact, three sets of experiments were cultured over multiple generations in different conditions, with one copepod culture at a temperature of 7 °C and another at 20 °C, both at salinity 15.

The third culture was performed at salinity 25 at 20 °C. All the experimental treatments were maintained under controlled conditions, with a 12 h/12 h light/day cycle in an acclimated environmental chamber in 2 L beakers for five generations [14,25,32,35]. A volume of 10 mL of Rhodomonas salina, a red microalga, was fed to E. affinis once every two days, with a concentration of 4 million cells. For each condition, ovigerous females were selected to start the generation, and these were separated after the release of eggs. Each generation was thereafter followed till the appearance of the next batch of ovigerous females. All the stages were carefully recorded and fixed accordingly for further analysis. From each generation of all culture conditions for the laboratory study, 20 to 40 ovigerous females were taken (

Table 2. Exposition of the different experimental conditions of E affinis (isolated from the Seine estuary) culture in the laboratory with the number of generations and number of total ovigerous females observed in each experimental condition.

Initial Experimental Conditions				Final Experimental Condition
Temperature	Salinity	No. of Generations	Total No. of Observations	Temperature	Salinity	No. of Generations	Total No. of Observations
7 °C	15	4	121	24 °C	15	5	112
20 °C	15	8	205	24 °C	15	5	196
20 °C	25	7	182	24 °C	25	5	135

). Similarly, from each in situ sample (field study), 10 to 30 ovigerous females were sorted depending on their availability.

The initial experimental conditions represented cold (winter like, 7 °C) and warm (summer like, 20 °C) conditions that are proven to be the two critical periods in the seasonal pattern of E. affinis population in the Seine estuary and Europe in general [36]. The final conditions of 24 °C corresponded to a warming scenario of +4 °C. The incorporation of two salinities (normal, 15 PSU, and stressful, 25 PSU) also corresponded to the climate scenario specific to the low-salinity gradient zone of the Seine estuary, the current habitat of the species [37]. The transfer from 7 °C to 24 °C was used to test the capacity of the species to survive a sudden severe heat shock and accordingly analyze the reproductive strategies of E. affinis in extreme conditions. Sudden heat shocks often impact several physiological functions, such as enzymatic activities and denaturation of proteins, including the alteration of membrane structures [38]. This treatment with extreme shock can be considered as a ‘negative control’ that depicts a significant distinction between the lower (7 °C) and higher temperature (24 °C), through the egg packing constraint.

In both the samples from the culture and field, every female sorted was observed under an inverted microscope, and the varied morphological and reproductive traits for this study were measured. The prosome length and width were measured, as detailed in [35], along with the consideration of the genital somite width. Particular attention was attributed to the length of the egg sac, which resembled a complete egg structure, with all the eggs inside the sac surrounded with the membrane (Figure 2). The maximum length was measured from the attachment part in the genital somite to the end of the sac (the limit of the membrane when it is visible) (Figure 2). The width measurement, in lateral view, was considered in the middle of the sac from one side limit of the membrane to the other side. The clutch size was counted by tearing the sac carefully, and 5 to 15 eggs per sac were measured [14,32]. All the measurements were realized by using ImageJ software, version 1.54m.

2.2. Volume Estimation

Both prosome and egg sac shapes were assimilated to an ellipsoid; consequently, their volumes were calculated by using Equation (1)

$$Xv={\displaystyle \frac{4}{3}}pi.\left({\displaystyle \frac{Xl}{2}}\right).{\left({\displaystyle \frac{Xw}{2}}\right)}^2$$

(1)

where Xv is prosome (or egg sac) volume (mm³), Xl is prosome (or egg sac) length (mm) and Xw is prosome (or egg sac) width (mm).

The number of eggs separated from the egg sac, which is the clutch size, was carefully counted and recorded for each condition [14,37]. After counting the number of eggs, the diameters of all eggs were measured in each condition. Eggs of E. affinis are spherical in shape and, thus, the egg volume was calculated from the average egg diameter [14,31].

The clutch volume (Cv, mm³) of each female was estimated by multiplying the total number of eggs [(clutch size (CS)] by the mean egg volume (Ev) considered as spherical of each female clutch by using Equation (2).

Cv = CS × Ev

(2)

2.3. Packing Constraint Estimation

To quantify the packing constraint, we suggest a simple index that we call “Packing Pressure Index” (PP index) by computing the difference between the observed values of the egg sac volume (ESv) and the clutch volume (Cv) using Equation (3).

PP index = ESv − Cv

(3)

when the PP index was negative, this was considered as a strong packing constraint. However, in order to consider the presence of the free volume between eggs in their distribution in the sac (knowing that all eggs without heavy packing constraints are spherical), we used two reference lines: y = x and y = 1.5x. When observed data were below the line y = x (i.e., ESv ≤ Cv), we considered that there was a strong packing constraint (SPC). When observed data were between the two lines (i.e., Cv < ESv ≤ 1.5Cv), we considered that there was an intermediate packing constraint (IPC) between clutch volume and sac volume; when observed data were above the y = 1.5x line (i.e., ESv > 1.5Cv), we considered that there was no packing constraint (NPC).

2.4. Data Preparation and Statistical Analyses

In this work, we used two different databases, one from the long-term multi-generation experiment that aimed to test the effect of global warming and a severe heat shock under different thermal and saline regimes [31,37] and another database from a collection of samples from different estuaries in Europe and North America. The methodology used to obtain the key morphological traits was the same for each sample of 20 to 40 ovigerous females (one sampling site on a single date or one cohort from a generation under the same experimental conditions) and all reproductive traits were obtained at the individual female level. Then, each database was analyzed separately.

In order to study the general trends between all reproductive traits selected in this study, the average values of each sample (in the field) or generation (in the lab) were used. Then, a linear regression analysis was performed, and the statistical significance was tested using Pearson correlation coefficient, with a threshold value of p < 0.05.

In order to apply the new Packing Constraints Index, individual data based on a single female were used, and to facilitate the visualization of these results, the egg sac volume was plotted against the clutch volume. Then, a linear regression analysis y = ax (intercept at the origin) was applied to each database (field and lab), and the statistical significance was tested using the Pearson correlation coefficient with a threshold value of p < 0.05. Although the data were highly dispersed, the slope of the linear regression was close to 2, which means that the estimated egg sac volume was, on average, double the real clutch volume (sum of the volumes of all eggs). Consequently, we arbitrarily added two additional reference lines. The first was y = x to represent a perfect situation, where the female used the exact volume of the sac (as estimated in this study) to pack all eggs. The second reference line was y = 1.5x that represents approximatively the median line between the regression line, with a slope ~2 and the previous line y = x. These lines allowed us to realize a classification of individual data to determine the percentage of three situations of packing constraints: no packing constraints (data above y = 1.5x), intermediate packing constraints (data between y = x and y = 1.5x) and data with strong packing constraints (data below y = x).

All statistical analyses and fittings were realized using the MATLAB software version 8.5.

3. Results

3.1. Relationship Between the Reproductive Effort and Morphological Reproductive Traits

The clutch volume was significantly correlated to the prosome volume in both conditions: in situ (Figure 3A) and laboratory (Figure 3B). The combination of all the data of prosome and clutch volume in all estuaries showed a low dispersion, around the linear regression lines under six experimental conditions (R² = 0.874, p < 0.0001) and high dispersion in situ (R² = 0.620, p < 0.0001) (Figure 3A). As expected, the egg sac volume showed a strong positive relationship with the clutch volume, in all E. affinis populations from the field (Figure 3C, R² = 0.817, p < 0.0001) and the laboratory (Figure 3D; R² = 0.929, p < 0.0001). For in situ data, the European populations were closer to the regression line than those from North America, whereas, for laboratory conditions, 80% of the data were situated within the confidence limits of the regression lines (Figure 3D). Since egg sac volume was a good proxy for clutch volume (Figure 3C,D), we logically obtained a good relationship with the prosome volume but with more dispersion for field data (R² = 0.524, p < 0.0001 for the field and R² = 0.841, p < 0.0001 for the laboratory). The egg sac volume had a close relationship with the genital somite at the point of attachment, which was confirmed by the results from the laboratory with R² = 0.820 (p < 0.0001) and was also statistically significant in the field data (R² = 0.323, p < 0.001) in spite of the high dispersion.

3.2. Egg Packing Constraint in Field and Laboratory Conditions

If we consider that the egg sac volume is the sum of the volumes of all individual eggs plus the free volume remaining in the sac, we can suppose that the packing of eggs by a female can be a function of clutch size, egg size and the constraints encountered. In ideal conditions, the egg sac volume should be greater than the clutch volume (reproductive effort) without any packing constraint.

The dispersions observed in field data, particularly in Figure 3C, could be due to the problem of the packing constraint. In fact, we observed three levels of egg packing in ovigerous females of E. affinis. Figure 4 illustrates these situations for sacs with similar clutch sizes (between 45 and 50 eggs per clutch). The egg sac dimensions (length and width) increased when the packing constraints decreased because of the appearance of a free space between eggs, as shown in Figure 4c. On the contrary, when the egg sac volume seems to be smaller than the clutch volume produced, the eggs are piled up and tight (Figure 4a). Between these two situations, the packing of eggs can present a good geometrical distribution, with more or less tightness between eggs (Figure 4b). In general, the strong packing constraint (SPC) can alter the spherical shape of the eggs.

In most cases, it is difficult to visually distinguish the level of packing constraint, and so to quantify its status in both field and laboratory conditions, we first plotted all individual ovigerous females in the plane ESv as a function of Cv (Figure 5). Then, we considered two reference lines, y = x and y = 1.5x. These two lines constitute the limits of the strong packing constraint (SPC) and the no packing constraint (NPC), respectively. Data plotted between the two lines were characterized by an intermediate level of packing constraint (IPC).

Both field (Figure 5A) and laboratory (Figure 5B) data showed a strong linear relationship between ESv and Cv with similar and high R² (0.721 and 0.70, respectively). On the basis of the slope values, the ESv is almost double Cv in all conditions (Figure 5). To identify the packing constraint, we should focus on the lower part of the data situated below the regression lines because the positively dispersed data has the highest number of packing constraints (NPC). The packing constraints (IPC and SPC) were more frequent in field data (Figure 5A) compared to laboratory data (Figure 5B), especially for the SPC. The data from the different estuaries were well mixed in the regression plane (Figure 5A), contrary to the laboratory data (Figure 5B), where we can easily distinguish the separations between the initial (in blue color palette) and final (in red color palette) experimental conditions. Under laboratory conditions, the IPC and SPC mainly correspond to data from high temperature and/or salinity (Figure 5B).

We computed the percentages of observed data from each constraint level by considering three classes, SPC, IPC and NPC.

Table 3. Percentages of the different case of packing in the different laboratory condition culture of E. affinis. The initial experimental conditions represented cold (winter like, 7 °C) and warm (summer like, 20 °C) conditions. The final conditions of 24 °C corresponded to a warming scenario of +4 °C during summer.

Packing Constraint (Types)	Initial Condition	Final Condition	Initial Condition	Final Condition	Initial Condition	Final Condition
	T7S15	T24S15	20TS15	T24S15	T20S25	T24S25
Strong packing constraint (%)	1.6	3.4	0.5	2	4.3	6.9
Intermediate packing constraint (%)	1.6	20.7	10.2	14	12.9	32.4
No packing constraint (%)	96.7	75.9	89.3	84.0	82.8	60.7

confirms that the percentage of packing constraints in the laboratory conditions (SPC and IPC classes) increased at the highest temperature of 24 °C, whereas the percentage of NPC decreased in all treatments. The higher salinity (25) compared to the optimal one (15) produced the highest percentages of packing constraints (SPC and IPC).

The distributions of the packing constraint class, in the different estuaries, are shown in Figure 6a, and they reveal clear inter-estuarine differences. In European estuaries, a north–south decreasing gradient in packing constraints appeared, giving the highest values (SPC = 32.35% and IPC = 35.29%) in Gironde and the lowest ones in the Scheldt (SPC = 2.44% and IPC = 9.76%). On the contrary, North American estuaries did not present a clear gradient, but the lowest packing constraint percentages were noticed in Chesapeake Bay (SPC = 4.17% and IPC = 8.33%), which were different from the St Lawrence estuary sites, showing a strong heterogeneity with a relatively high packing constraint.

The classification of all sampled estuaries/sites based on their relative distributions in the selected three classes of packing constraints and using Chi-square distance gave the dendrogram in Figure 6b. This later opposed two different groups of estuaries, one composed of three sites in Europe (Elbe, Scheldt, and Seine) and two sites in North America (Berthier Sur Mer (SL) and Chesapeake Bay), and the second one was composed of two sites from Europe (Loire and Gironde) and three sites from St Lawrence in North America (Isle Verte, St Jean Port Joli, and Montmagny). In the first group, the NPC was always higher than 60%, and the SPC never reached 18%, where the lowest percentage was null in the Elbe estuary. In the other group, the NPC never reached 50%, while the SPC was always greater than 23%. The site of Montmagny (SL3) showed the lowest similarity in the second group (Figure 6b) because it contained the lowest NPC percentage in all estuaries/sites (13.33%) and the highest percentages of packing constraints SPC and IPC (33.33% and 53.33%, respectively).

In order to understand the effects of the egg packing constraints on the relationship between the ESv and the key reproductive traits (Cv, Ev, Pv and GSw), we selected the estuaries showing more than 10% of SPC. This comprised three estuaries in Europe (Seine, Loire, and Gironde) and four sites in the St Lawrence estuary in North America (

Table 4. Table showing the relationships with the egg sac for estuaries that present more than 10% of strong packing constraint.

Continent		Europe			(North) America
Estuaries		S	L	G	SL1	SL2	SL3	SL4
ESv vs. Cv	SPC	0.88	0.90	0.95	0.67	0.76	0.96	0.97
	IPC	0.98	0.96	0.996	0.98	0.97	0.88	0.97
	NPC	0.86	0.87	0.83	0.87	0.86	-	0.53
ESv vs. Ev	SPC	10−5	0.03	0.62	0.07	0.73	0.35	0.21
	IPC	0.04	0.01	0.003	0.27	0.01	0.13	0.54
	NPC	0.08	0.15	0.25	0.34	0.05	-	0.18
ESv vs. Pv	SPC	0.06	0.53	0.004	0.21	0.60	0.71	0.22
	IPC	0.10	0.70	0.21	0.65	0.10	0.14	0.49
	NPC	0.44	0.73	0.12	0.66	0.13	0.13	0.27
ESv vs. GSw	SPC	0.06	0.15	0.007	0.31	0.33	0.002	0.03
	IPC	0.20	0.06	0.52	0.15	0.26	0.04	0.94
	NPC	0.38	0.32	0.04	0.11	0.05	0.19	0.04
SPC (%)		14.65	25.00	32.35	25.00	23.00	17.39	33.33
IPC (%)		21.21	28.13	35.29	36.11	27.66	53.33	21.74
NPC (%)		64.14	46.88	32.35	38.89	48.94	13.33	60.87

). The correlations between ESv and the other selected traits (Cv, Ev and Pv) were calculated in each class of egg packing constraint (SPC, IPC and NPC). For all combinations, the ESv was highly correlated with the clutch volume (Table 4). The highest correlations were obtained for the IPC conditions [except for Montmagny (SL3)] followed by SPC that showed a stronger correlation for ESv vs Cv compared to ESv and Ev, which was generally weak. In fact, in SPC, only two estuaries (G and SL2) showed highly significant correlations (Table 4), with no strong correlations obtained for IPC conditions. The Seine estuary along with Loire and SL1 showed highly significant correlations for NPC. The correlation between ESv and Pv was highly significant, in the case of NPC (except for Gironde and SL3, Table 4). For IPC, the relationship was highly significant for two estuaries, Loire and SL3. For SPC, in addition to Loire, SL2 was also highly correlated.

For the selected estuaries of Europe, the correlations between ESv and GSw in the Seine and Loire populations were highly significant in the case of NPC, whereas the Gironde population showed this relationship as highly significant in the case of IPC. For the St Lawrence sites, except for SL4, which showed significant correlation in IPC, no relationship between the ESv with the genital somite was observed.

For the remaining estuaries (Elbe, Scheldt, and Chesapeake Bay), where SPC was very low, the previous correlations in NPC were valid for ESv vs. Cv. The ESv of ovigerous females from the Chesapeake Bay was highly correlated to the Ev (R² = 0.38) but showed a low correlation (R² = 0.08, p < 0.05) with the GSw. On the contrary, the other European estuaries (Elbe and Scheldt) showed high correlations (p < 0.005) with GSw (R² = 0.58 and R² = 0.29, respectively).

By considering all estuaries in NPC, it seems that only the European populations showed a functional relationship between the genital somite and the size of the egg sac.

3.3. Relationship Between Packing Pressure Index and Clutch Size

In order to understand the possible mechanisms affecting the egg packing constraints, we selected two extreme situations from the different experimentally controlled conditions. From Table 3, we identified the highest percentages of NPC in the condition of T7S15 (96.7%), which is considered as a control without any pressure of packing. The highest percentages of SPC were observed at the highest temperature and the highest salinity, but the total number of females was relatively low per single condition (Table 3). Consequently, in order to test the relationship between the CS and PP index of SPC, the individual data were aggregated. Figure 7 shows a significant linear relationship between the CS and PP index with two opposed trends, a positive slope for the T7S15 (i.e., NPC) condition and a negative slope in the presence of SPC. The packing constraints seemed to be independent of the clutch size. In fact, at T7S15, females were able to produce more than 80 eggs per clutch without showing any packing constraint. In contrast, for the strong packing constraint situation, even if the clutch size was smaller (maximum CS of 43 eggs) due to the high temperature and/or salinity, a negative trend was observed.

4. Discussion

4.1. Presence of True Egg Sac with External Membrane in Eurytemora affinis

Our study was based on a high number of detailed examinations of ovigerous females of the cryptic species complex of the calanoid copepod E. affinis, confirming the presence of a true egg sac with external membranes embracing the eggs. This observation was valid for 10 populations from different estuaries in Europe and North America, which increases the case studies on the presence of an egg sac membrane among Copepoda Calanoida [1,19,20]. This external membrane is of great importance in the reproduction of these egg-bearing copepods due to its protective and nutritive role to the eggs [39].

Generally, the studies of the egg sac and its membrane are very scarce, with only one study, suggesting the mechanism of egg membrane formation for the freshwater calanoid copepod Heliodiaptomus viduus [39]. This work described the presence of a secretory gland of the oviduct, responsible for the production of the required materials necessary to build the external sac membrane. Since E. affinis is a brackish species able to live in the low-salinity zone of estuaries [36] and even invades freshwater habitats [40], we can hypothesize that its egg sac membrane formation could result from similar physiological mechanisms. According to Altaff and Chandran [39], the oviduct gland produces an elastic sac that envelopes the oocytes until their full development; then, the elastic sac detaches from the oviduct, bringing the eggs to the genital somite, ready for internal fertilization. As soon as eggs are fertilized, the elastic sac transforms to an external sac, enclosing the eggs with a membrane [39]. In E. affinis, this egg sac is clearly attached to the genital somite after its formation (Figure 2). However, measurements of packing pressure can be influenced by the attributes of the egg clutch, such as the movement and physical properties between fertilized and unfertilized eggs. Fertilized eggs often have a different structure, such as a developing embryo compared to unfertilized eggs, which potentially alter their shape, rigidity, and their interaction with each other in stressed conditions. As eggs develop, their internal composition and structural properties change. For example, in fertilized eggs, the development of the embryo could lead to differences in the mechanical properties of the egg (e.g., changes in the elasticity or density of the egg’s contents). These variations influence the amount of pressure the eggs exert on each other within the clutch. In addition, the shell thickness, membrane integrity, and internal contents (like the yolk and albumen) can also vary. Eggs that are more or less rigid or that have different sizes might compress or adjust in the clutch differently, influencing the overall packing pressure. If the eggs are disturbed by environmental factors, animal behavior, or vibrations, this could alter their arrangement in the clutch and change the pressure distribution between eggs. Additionally, eggs that move relative to each other might exert different forces than those that remain stationary.

4.2. Packing Constraint in E. affinis Populations

The internal origin of the egg sac, composed of the membrane and a fixed number of spherical eggs, requires accurately scheduled steps that E. affinis females have to optimize. In fact, when the external membrane reaches the last steps before extrusion, only its size and elasticity delimit its maximum volume. However, this elasticity should have some limit, and the female will end the reproductive cycle by packing the eggs into the sac. This last step may lead to packing constraints when the reproductive effort (number of eggs per clutch × egg volume) is much higher than the maximal capacity (volume) of the external sac.

In order to track these situations in the cryptic species complex of E. affinis, we suggested the use of a simple index called the Packing Pressure Index (PP index), which considers the difference between the final egg sac volume and the reproductive effort (clutch volume). This index and the reference lines of the packing constraints (see Figure 5) permit the separation of the different patterns of packing constraints, which are SPC, IPC and NPC. Contrary to the clear separation between European and North American populations of E. affinis based on the trade-offs of their reproductive effort, the packing constraints analyses separated two groups of estuaries by mixing transatlantic populations (Figure 6). The first group of estuaries is composed of the southern populations in Europe (Gironde and Loire), and the northern ones in St Lawrence estuary (Isle verte, St Jean Port Joli and Montmagny) are characterized by a high SPC and/or IPC. However, the possibility that strong packing pressure is a by-product of other aspects of reproductive failure or abnormality is an important consideration in understanding egg clutch dynamics for future studies. The second group, showing the highest percentages of NPC, assembled the at the southern site in St Lawrence (Bertier sur Mer), the Chesapeake Bay, representing the most southern estuary in this study, and three northern European estuaries (Seine, Scheldt, and Elbe). Our results underlined that the packing constraint situation of E. affinis ovigerous females was primarily dependent on the geographical location of the estuary rather than the genetic similarities (North American clades vs. European clade). Indeed, in Europe, a clear positive north–south gradient in the distribution of the packing constraint cases was observed, with the highest SPC and NPC in southern and northern estuaries, respectively. The North American estuaries considered in this study showed that the gradient in packing constraint appeared to be opposed to the European ones. In this case, the bigger dimensions of the St Lawrence estuary and the Chesapeake Bay than the European estuaries globally contrasted in their packing constraints. E. affinis is a complex species, and its fine morphological and/or phylogenetic description has already shown the presence of a new species, E. carolleeae [41], as well as sub-populations in European Waters [42]. For sake of simplicity, and because fine morphological data were not available in our database, we deliberately used the term ‘cryptic Eurytemora species complex’. We believe, in all cases, that E. affinis was the dominant species in the samples, and the presence of specimens belonging to E. carlleeae did not seem to modify our conclusion. In fact, Figure 6b, based on packing constraint indices, showed, for example, that the Gironde estuary in France was very similar to the Chesapeake Bay in the USA, where E. carolleeae could be found. We can logically hypothesize that the packing constraints for very close species did not seem to be the result of a clear evolutionary pattern leading to the speciation within the cryptic Eurytemora species complex. On the contrary, we believe that other abiotic factors, such as salinity or water density, could affect the packing constraints, and this should be carefully considered in future studies.

This experimental study based on a representative population from the Seine estuary revealed that the SPC and IPC situations increased with an increase in temperature and/or salinity. This finding confirms the observed north–south gradient in Europe that may not reflect only a thermal gradient but also a clear difference in the quality of the habitat of E. affinis. The contrasting situations between Gironde and Scheldt observed in this study (Figure 6) reinforced this statement. In fact, Gironde is one of the most turbid estuaries situated in the southern part of Europe, where ovigerous females of E. affinis presented a small size and low egg production, particularly in recent years [14,43], whereas the Scheldt estuary presented a suitable habitat for E. affinis (with larger females and higher clutch sizes), particularly after the recent restorative actions improving the water quality [44]. In North American estuaries, the observed differences in the repartition of the packing constraint classes in the cryptic Eurytemora species complex ovigerous females, especially for St Lawrence, could lead to highly variable habitats in the context of the salinity gradient [29]. The four sites considered in this study contain one salt marsh (Isle verte) and three sites affected differently by the tidal regime, where two clades of E. affinis could co-occur [29]. Ovigerous females of E. affinis sampled at Montmagny showed the highest constraints in egg packing, which was opposed to the site of Berthier sur Mer, in spite of their close locations in the oligohaline zone. In addition to the genetic heterogeneity between E. affinis populations in the St Lawrence estuary [29], the differences between habitats, as pointed out by the packing constraint analysis, should be considered. The packing constraint patterns in the ovigerous females from the Chesapeake Bay were similar to those of the Scheldt estuary, indicating that it is a favorable habitat for E. affinis.

4.3. Effect of the Packing Constraint on the Relationship Between ESv and Key Morphological and Reproductive Traits

The reproductive effort (i.e., Cv) and the ESv were positively correlated to the body size of E. affinis ovigerous females from both field and laboratory investigations (Figure 3). In general, the correlations between ESv and Cv were highly significant in all packing constraint situations, with the highest R² often observed in IPC situations (see Table 4). Therefore, we can consider the ESv as a proxy for the reproductive effort of egg-bearing copepods that can easily be measured. The other relationships between ESv and morphological traits were influenced by the packing constraint situation. For example, the ESv and Pv were better correlated in the NPC situation, which suggested that the reproductive energy (leading to a clutch and a sac) was well scaled with female size in this situation. The experimental results using E. affinis from the Seine estuary revealed that the ESv is linearly related to the genital somite width. When all data from the field were combined, this relationship was significant but with a high dispersion due, principally, to the North American populations (data not shown) that did not show any significant correlation when considered alone (R² = 0.14; p = 0.12). In fact, when only European populations were used, a strong correlation was obtained (R² = 0.70; p < 0.0001). Therefore, this relationship between ESv and GSw seems to be valid only for European populations in the NPC situation. Gironde was the only estuary where the correlation between ESv and GSw was only significant in the IPC situation (Table 4). Moreover, Gironde was the only European population showing a strong correlation between ESv and Ev in SPC conditions (R² = 0.62; p < 0.0001). This relationship was also valid in NPC situations for Scheldt, Seine, and Loire in Europe and for Isle verte and Chesapeake Bay in North America. Beck and Beck (2005) [23] used a numerical algorithm to optimize the quantity of eggs that could be packed in a body cavity of a given shape and size. They hypothesized that the relationship between allocation to reproduction (clutch size times maximum egg size) and the egg volume should be positively linearly related in log–log plots, with or without packing constraints. We showed, in this study, that E. affinis can exhibit a range of packing constraint situations, and, therefore, we cannot assign it to a single reproductive strategy (with or without packing constraints). In absolute terms, all egg-bearing copepods should face the packing constraints because the reproductive effort could not evolve independently of the egg size [23]. But the positive relationships between ESv and Ev in SPC situations were only observed in Gironde and St Jean Port Joli populations. For the European case, this is probably due to the low number of eggs per clutch.

4.4. Importance of an Egg Sac in Egg-Bearing Copepods: Generalities with Egg Packing Constraint

The present study is the first to deeply analyze the packing constraints in a single species of Copepoda Calanoida. The conclusions of this study confirmed that the packing constraints in Calanoida occur and could have an ecological meaning. The relationship between the reproductive effort, the packing constraints and egg volume should be carefully addressed in future studies. In fact, the egg sac is of great importance in the life history of egg-bearing Calanoida being involved in the survival of the species by ensuring the egg carrying [3], increasing the vulnerability to predation [45] and also the adaptive success of the invasion of freshwater habitats [46]. Therefore, egg-bearing Copepoda facing these opposed selection pressures evolved to produce either a single (Calanoida and Harpacticoida) or two egg sacs (Cyclopoida) that differ in shape and in position [3]. The egg packing constraints, ignored in most copepod studies, could reflect the copepod responses to the environmental conditions. The only study considering egg packing constraints in copepods was performed by Beck and Beck (2005) [23]. The database used originated from Caley et al. (2001) [47], which, in turn, was mostly based on Poulin’s work [48], containing a compiled dataset on free-living (broadcast spawners and egg bearing) and parasitic copepods. Moreover, the high dispersion of their datasets made it difficult to generalize their application to Calanoida. However, we believe that the packing constraint algorithm [23] could be adapted to Copepoda but at lower taxonomic levels with ecologically similar species.

4.5. Ecological Significance of Packing Constraint

The packing constraints observed in the field were much higher than those observed under controlled conditions, where the high number of external factors that could affect the packing constraints could not be accurately controlled. Consequently, we considered two opposing situations in the experiments to compare the SPC and the NPC by plotting the clutch size as a function of the PP index (see Figure 7). In the T7S15 condition, where we found the highest percentage of NPC, the relation with the clutch size followed a positive trend, meaning that the greater the clutch size, the more the PP index increased positively (i.e., increasing the NPC situation). In contrast, for the case of SPC that we observed at a salinity of 25 and temperature of 24 °C, the clutch size was negatively correlated with the PP index, which means that the greater the clutch size, the higher the constraint. These patterns could be explained by the fact that at the lower temperature, the fecundity and survival were high, and the growth was slow, so the egg sac size always matched the clutch size produced. However, in stressful conditions, when the female invested more energy in her survival, the fecundity was usually lower. Furthermore, the water density was higher at low temperature and could offer an advantage to sustain better-quality females and their egg sacs. Indeed, when the female produced a bigger clutch size compared to the “average situation”, the packing pressure appeared because the egg sac size remained constant. This suggests that the reproductive effort and the ESv are not usually in equilibrium. The female can receive an environmental signal that allows her to prepare a certain reproductive effort, but, during egg laying, the female can face some perturbations from different sources of stress that may favor the appearance of packing constraints. The external egg sac carried by ovigerous females resulted from two internal reproductive mechanisms, leading to simultaneous or sequential steps of membrane and egg clutch (oocyte maturation, egg fertilization and their release) formations. The quantity of material used to produce the elastic internal sac should be correlated to the clutch size in order to avoid any disequilibrium between the number of fertilized eggs to be extruded and the carrying capacity of the membrane (without reaching the limit of its elasticity). In the situation when the equilibrium between egg clutch and membrane sizes was reached, NPC was expected. However, for any internal or external reasons giving higher egg production, approaching the limit of the distortion capacity of the elastic membrane, the packing constraint pressure was enhanced, and the eggs appeared to be piled up tight in the sac. In SPC, it seems that a clear desynchronization exists between the signal, leading to the egg membrane formation and the allocation of the reproductive effort.

All abiotic conditions encountered by not only the female but also the preceding generation (via maternal control) could determine its fitness, including reproductive performance. However, in field conditions, it is almost impossible to determine the age of copepods with precision. Also, the quantity of food represented by chlorophyll a or a micro-plankton composition [36] may not be the best indicators of the food items available to Calanoida that are also omnivorous. However, pigment content remains the most used indicator for its common and relatively easy quantification. As a consequence, we developed several morphometric analyses based mainly on body size, body shape and the reproductive traits of copepod females to build additional indicators of stress or individual conditions similar to the protocol adapted for other aquatic model organisms, such as fishes. For example, Souissi and Souissi (2021) [24] used a simple clutch size vs. prosome length ratio of E. affinis to suggest an integrative habitat quality index. It turns out that this index discriminated against the situations when copepods encountered unfavorable conditions in the field and confirmed that most samples from the European estuaries fall within the normal trend of CS vs. PL predictions.

These different egg packing constraint situations can be observed in the field and can indicate some external sources of stress or perturbation of the reproductive cycle of ovigerous females. This study showed that the packing constraints were site/estuary-specific and did not show any similarity to the genetic structure of the considered populations/clades but concerned the environment faced by the complex and cryptic populations of Eurytemora species.

Our results suggest that the packing constraints analysis can be used as an indicator for the habitat quality of E. affinis populations within and between different estuaries. This hypothesis can be tested experimentally by exposing different populations of E. affinis to several sources of external perturbations and comparing both the outer and inner egg membrane structures to have significant implications for egg survival, hatching success, and reproductive strategies in copepods. In addition to natural factors that could affect the reproductive strategy of E. affinis and the packing constraints in the field, the presence of pollutants and complex cocktails of chemicals in estuarine waters should also be considered in future studies. When examining packing constraints, future studies should be cautious with the egg types, as E. affinis is known to produce subitaneous and resting eggs that might differ not only in size but also in the robustness of the chorion [49].