Abstract

Many ectotherms grow more slowly but mature at a larger size in colder environments, according to the pattern called the temperature-size rule. Thermal variation of cell size is suspected to be inherent in the origin of this pattern, but empirical testing of this hypothesis has been conducted in only a few taxa. In the laboratory, we reared two subspecies of the land snail Cornu aspersum (formerly Helix aspersa), C. a. aspersum and C. a. maximum, at two thermal regimes (15 and 20 °C), aiming to examine the relationship between cell size and adult mass across temperatures. The warmer environment led to larger adult mass in both taxa, contrary to the temperature-size rule. Using histological techniques, we found that snails grown at the warmer temperature produced heavier shells, smaller muscle and epithelial cells, but larger hepatopancreatic cells and nuclei. These results strongly suggest that the temperature-size rule for body size cannot be explained by a simple consequence of thermal sensitivity of cell size. Focusing on the patterns between the two taxa, we found that aspersum grew more slowly, with faster metabolic rates and smaller mass at the end of the experiment, when compared with maximum, regardless of temperature. The subspecies aspersum evolved larger muscular and epithelial cells than maximum, but both had a similar size of hepatopancreatic cells and a similar shell mass. Interestingly, when compared across a range of small body masses, aspersum had a heavier hepatopancreas than maximum. We propose that the evolutionary divergence between the two subspecies might involve a trade-off between growth and hepatopancreatic functions, but not a trade-off caused by the costs of shell production and maintenance of cell membranes.

INTRODUCTION

In many ectotherms, body size varies according to the environmental temperature. Low developmental temperatures typically cause ectotherms to grow slowly, to postpone maturation and to become reproductively active at a larger body mass, a plastic response called the temperature-size rule (Atkinson, 1994; Forster, Hirst & Atkinson, 2012; Klok & Harrison, 2013). Among populations, body size diverges genetically such that larger individuals occur in colder environments, a pattern called Bergmann's rule (Zwaan et al., 2000; Ashton, 2002; Angilletta et al., 2004). Thermal sensitivity of cell size might be inherent in the origin of these patterns (van Voorhies, 1996), but studies in this field have usually considered either a single cell type (Arendt, 2007; Czarnoleski, Dragosz-Kluska & Angilletta, 2015) or eutelic organisms, which have constant cell numbers (van Voorhies, 1996; Walczyńska et al., 2015).

For many species, body size has clear evolutionary consequences: large individuals can produce more and/or larger offspring and suffer lower predation risk, whereas prolonged growth or increased foraging before maturation decreases the chance of reaching maturity (Kozłowski, 1992; Stearns, 1992). The evolutionary consequences of cell size, however, are much less explicit. An emerging theory of optimal cell size (Szarski, 1983; Woods, 1999; Kozlowski, Konarzewski & Gawelczyk, 2003; Atkinson, Morley & Hughes, 2006; Czarnoleski, et al., 2015) suggests that a balance between the need to meet metabolic demands and the need to conserve energy determines the optimal size of cells. Small cells produce a large area of cell membranes relative to volume. Consequently, organisms with smaller cells waste more energy because cell membranes require energy for phospholipid turnover and to generate electrochemical gradients. The cost associated with the maintenance of small cells is expected to favour large cells in cold ectotherms (Atkinson et al., 2006; Czarnoleski et al., 2013, 2015). Nevertheless, organs with small cells should metabolize more rapidly because of the greater surface area of membranes for transport, shorter distances for diffusion and more nuclei (genomes) per volume available for transcript production. The advantages of small cells are expected to select warm ectotherms for building their body of small cells (Atkinson et al., 2006; Czarnoleski et al., 2013, 2015).

The garden snail, Cornu aspersum (Müller, 1774) (formerly Helix aspersa), is a terrestrial pulmonate gastropod of the family Helicidae. This species has two infraspecific taxa, C. aspersum aspersum and C. a. maximum (referred to as CAA and CAM hereafter). The split into CAA and CAM is estimated to have occurred in the mid to late Pliocene, and the two taxa now live sympatrically in parts of North Africa (Guiller et al., 2001, 2006). CAA also extends north to the temperate regions of western Europe and the Black Sea region and, as a result of anthropogenic activity, it now occurs in the Americas, Australia, New Zealand and South Africa (Guiller et al., 2012). We expect that comparing the thermal reaction norms between these two taxa may provide an important physiological perspective for understanding how an ectotherm can adapt to heterogeneous environments across latitudes.

The main goal of this study is to gain an insight into the cellular basis of thermal plasticity of body size in ectotherms. For this purpose, we reared CAA and CAM snails at two temperatures, and examined the thermal plasticity of growth and final body size. We checked whether the origin of CAA and CAM involved differentiation of developmental responses to thermal conditions. We also measured the cells of adults in the foot epithelium, the locomotory muscles and the hepatopancreas to investigate how cell size contributes to the thermal plasticity of body size and to the size differences that have evolved between the subspecies. We measured the rate of metabolism and examined its links to differences in cell size between snails. Following the concept of optimal cell size, we expected the warm snails to develop small cells and the snails with small cells to have a faster metabolism. Moreover, if large cells help to reduce metabolic costs, the subspecies that extends to high-latitude environments (CAA) should be characterized by larger cells than the more restricted subspecies (CAM).

MATERIAL AND METHODS

Growth pattern

A total of 1200 snails of CAA and CAM, 4 d of age, were obtained from the National Research Institute of Animal Production (NRIAP, Balice, Poland). The stock of snails in NRIAP originates from breeding snail farms of the French National Institute for Agricultural Research and its establishment was supported by Polish-French collaborative research programs. From each taxon, half the individuals were kept at ambient temperature of either 15 or 20 °C in thermal chambers at the Institute of Environmental Sciences (Jagiellonian University, Kraków, Poland) for c. 13 months under a photoperiod of 16:8 LD, without a hibernation period. Note that the chosen temperatures are below the upper thermal critical zone of Cornu aspersum (Gaitan-Espitia et al., 2013). For each temperature and subspecies, snails were kept in a single Plexiglas container (50 × 50 × 15 cm) and were fed a standard diet. For logistical reasons, we did not replicate containers in experimental groups, which potentially confounds the effects of a container and thermal treatment. The food was a powdered mixture of plants, minerals and vitamins, enriched with dry soil and an additive of CaCO3 (Łysak, Mach-Paluszkiewicz & Ligaszewski, 2001).

To monitor growth, snails from each box were regularly counted and the total snail mass per box was recorded on an electronic balance (accuracy 0.001 g) until the end point of the experiment, when snails reached 400 d of age. Given that few snails remained at this age, we assume that at the end of our experiment snails were very close to their physiological maximum lifespan under constant conditions and prevented hibernation. The mean individual body mass in a box was calculated for each time interval, which was usually equal to 7 d, but for logistical reasons, it ranged from 3 to 11 d. The last three intervals, when growth was considerably slower, were 15, 27 and 31 d. Snail number decreased over time due to mortality and sampling of the snails for respiration measurements. The growth pattern was described by the two-parameter von Bertalanffy's equation, following the methods of Czarnoleski et al. (2003). Equation fitting and all other statistical analyses were performed in Statistica vs 10 (StatSoft, Poland).

At the end of the experiment, snails were weighed individually to the nearest 0.001 g with and without the shell. After fixation of the soft parts in Bouin's solution, the hepatopancreas of each snail was dissected and weighed to the nearest 0.00001 g. With the help of histological methods, we assessed maturation state of the snails and the cell sizes in tissues (see below). In total, we examined the cell size and maturation state in 12 CAM snails at each temperature, 12 CAA snails at 20 °C and 11 CAA snails at 15 °C.

We applied a general linear model (GLM) to body mass data, with the snail taxon and the thermal treatment as the fixed-effect factors to examine their effects on body mass. We used a similar GLM, with whole-body mass as a numerical covariate, to compare shell mass (calculated as the difference between the whole-body mass and mass of the soft parts) and hepatopancreatic mass among the snails. Whole-body mass, shell mass and hepatopancreas mass were transformed with decimal logarithms prior to the analysis.

Reproductive state and cell size

Snails that survived to the end of the growth experiment were further studied with histological methods. To assess the maturation of the snails, we dissected gonads and gradually dehydrated them in an ethanol series (70, 80, 90, 96%), cleared them in ST Ultra (Leica, Germany) and embedded them in Paraplast Plus (Leica). Serial 4-µm sections were cut with a motorized rotary microtome Hyrax M55 (Zeiss, Germany), stained progressively with Ehrlich's haematoxylin (Carl Roth, Germany) and counterstained with a 1% solution of eosin Y in ethanol (Analab, Poland). After dehydration in 96% ethanol, the sections were cleared in ST Ultra (Leica) and mounted with CV Ultra (Leica). Slides were examined under a light microscope (Eclipse 80i, Nikon, Japan) to assess the developmental phase of the oocytes and the spermatogenic cells in the reproductive follicles. Snails were regarded as reproductively mature if late vitellogenic or ovulated oocytes or spermatozoa were observed or if reproductive follicles were spent.

To examine cell size, we dissected the hepatopancreas and a rear fragment of the foot with locomotory muscles and external epithelium. Tissue samples were gradually dehydrated in an ethanol series (50, 60, 70, 80, 90, 96%), transferred to butyl alcohol, cleared in chloroform and, finally, embedded in paraffin wax (Avantor Performance Materials, Poland). Serial 5-μm sections were cut with a motorized rotary microtome RM 2155. Cross sections of the foot were stained progressively with Ehrlich's haematoxylin and counterstained with water eosin Y (Fluka, Germany). Sections of the hepatopancreas were stained regressively with Heidenhein's haematoxylin (Fluka). Following dehydration in 96% ethanol and acetone, the sections were cleared in xylene and mounted on slides with DPX (Avantor Performance Materials).

We took photographs of the hepatopancreases and the foot muscles with a light microscope IX 71 (Olympus, Japan) equipped with a digital DP 70 camera (Olympus). Epithelial tissue in the foot was scanned with a BX 61 VS virtual microscopy system equipped with an XC10 camera and VS ASW FL 2.3 software (Olympus) and then bitmaps were produced with OlyVIA image viewer (Olympus).

We used AnalySIS vs 3.2 software (Olympus) to measure the digestive cells and their nuclei in the hepatopancreas and the locomotory muscles in the foot. Hepatopancreatic cells and nuclei were outlined and their area (µm2) was calculated; in dikaryotic cells, the total area of the two nuclei was calculated. The same technique was applied to measure the area of the locomotory muscle, which was cut in a transverse plane. Another set of muscle cells, which was cut longitudinally, had its diameter measured (µm). We used ImageJ software (National Institutes of Health, USA) to measure the epithelial cells in the foot. The cell membrane was not visible in epithelial cells; therefore, to assess cell size of this type of cells (µm), we identified groups of 2–9 adjacent nuclei that were aligned linearly in transects and measured the distance between the nuclei at two ends of each transect. For each individual snail, we measured 50 cells and 50 nuclei in the hepatopancreas, 50 muscle cells in their transverse section, another 50 muscle cells in their longitudinal section and 23–111 transects with epithelial cells.

Data on the area of the hepatopancreas cells and their nuclei were used to calculate a karyoplasmic ratio in the hepatopancreatic cells (the ratio of nucleus size to cell size). The mean karyoplasmic ratio for each snail was calculated from the ratios of individual cells. Data on the size of the cells in the muscles and the hepatopancreas were transformed with decimal logarithms and their mean values were calculated for each individual snail. To estimate the mean size of the epithelial cells in snails, we calculated the total length of the cell groups for each individual snail and divided it by the total number of nuclei in all groups. This way, each snail was characterized by four measures of cell size. We performed principal component analysis (PCA) on these data to produce integrated measures of cell size in the studied cell types (values of principal components). These measures were further analysed with GLM to compare cell size in the two subspecies of snails at two thermal treatments (fixed factors). The same analysis was used to compare the karyoplasmic ratio of the hepatopancreatic cells.

Standard metabolic rate

At age 136–150 d, 30 snails involved in the growth experiment were randomly sampled from each box for measurements of respiration rate. Snails were starved for 48 h before respiration measurements and weighed to the nearest 0.001 g immediately before measurements were made. Individual snails were placed in 1-l glass containers and enclosed in 5-mm mesh material to reduce their activity. To further reduce snail activity during measurements, snails were exposed to constant dim light. To provide humidity, the containers were supplemented with an open Eppendorf tube filled with distilled water. Temperature during respiration measurements was set to the respective snail thermal treatment. The total production of CO2 over 36 h (µl) was measured using a closed-system Micro-Oxymax respirometer (Columbus Instruments, USA) and this measurement was converted to average hourly production. A GLM was applied to log10-transformed hourly CO2 production data with the whole-body mass as a covariate and taxon and temperature as fixed-effect factors.

RESULTS

Growth and maturation

Warm snails grew faster than cold snails (Fig. 1). At the end of the growth experiment (age 400 d), warm snails had on average a larger mass than cold snails (Table 1), conflicting with the temperature-size rule. When compared at a given temperature, CAA snails had a smaller final body mass than CAM snails. The nonsignificant interaction between the snail subspecies and the temperature shows that the magnitude of thermal changes in body size was similar in each subspecies (Table 1). At the end of the experiment, most snails showed clear signs of reproductive activity: 11/11 cold CAA snails were mature, 11/12 warm CAA snails were mature (gonadal development was not assessed in one snail), 9/12 cold CAM snails were mature (gonadal development of one snail was not assessed; two snails had previtellogenic oocytes) and 11/12 warm CAM snails were mature (one snail had previtellogenic oocytes).

Table 1.

Statistics for the GLMs of snail final body mass (age 400 d), cell size measurements (PC1 and PC2, see Table 3) and karyoplasmic ratio (nucleus size to cell size ratio) in hepatopancreas cells.

Dependent variable Effect F1, 43 P 
log10 body mass Subspecies 9.54 0.004 
Temperature 7.94 0.007 
Subspecies × temperature 0.02 0.88 
PC1 Subspecies 6.04 0.02 
Temperature 5.03 0.03 
Subspecies × temperature 0.03 0.86 
PC2 Subspecies 0.24 0.62 
Temperature 4.36 0.04 
Subspecies × temperature 1.0 0.32 
Karyoplasmic ratio Subspecies 1.42 0.24 
Temperature 0.53 0.47 
Subspecies × temperature 0.54 0.47 
Dependent variable Effect F1, 43 P 
log10 body mass Subspecies 9.54 0.004 
Temperature 7.94 0.007 
Subspecies × temperature 0.02 0.88 
PC1 Subspecies 6.04 0.02 
Temperature 5.03 0.03 
Subspecies × temperature 0.03 0.86 
PC2 Subspecies 0.24 0.62 
Temperature 4.36 0.04 
Subspecies × temperature 1.0 0.32 
Karyoplasmic ratio Subspecies 1.42 0.24 
Temperature 0.53 0.47 
Subspecies × temperature 0.54 0.47 

Fixed factors: snail subspecies (Cornu aspersum aspersum and C. a. maximum) and developmental temperature (15 or 20 °C). Scores of principal components (PC1 and PC2) are an integrated measure of cell sizes in three different tissues (see Table 3 for details).

Figure 1.

Snails of Cornu aspersum aspersum attained a smaller body mass than snails of C. a. maximum. Both subspecies developed smaller mass in the warm environment. Curves are von Bertalanffy's growth equations fitted to data of the mean body mass.

Figure 1.

Snails of Cornu aspersum aspersum attained a smaller body mass than snails of C. a. maximum. Both subspecies developed smaller mass in the warm environment. Curves are von Bertalanffy's growth equations fitted to data of the mean body mass.

At the end of the experiment, shell mass did not differ between taxa if compared at a common body mass, but the warm snails generally had 33% heavier shells than the cold snails (Table 2). The mass of the hepatopancreas increased with the body mass, and the effects of body mass interacted with thermal treatment and with snail subspecies (Table 2). At a smaller body mass (Fig. 2A: arrow M), CAA snails had a 32% (warm) or 13% (cold) heavier hepatopancreas than CAM snails. At a larger body mass (Fig. 2A: arrow L), the mass of the hepatopancreas was 10% (warm) or 11% (cold) smaller in CAA compared with CAM.

Table 2.

Statistics for GLMs of shell mass and the mass of the hepatopancreas of Cornu aspersum snails (age 400 d). Fixed factors: snail subspecies (C. a. aspersum and C. a. maximum) and developmental temperature (15 or 20 °C); log10 body mass is a covariate.

 Log10 shell mass
 
Log10 hepatopancreas mass
 
F1, 39 P F1, 39 P 
Subspecies 2.17 0.15 13.66 0.0007 
Temperature 8.78 0.005 11.27 0.002 
body mass 241.69 0.000001 243.89 0.000001 
Subspecies × Temperature 0.22 0.64 4.24 0.046 
Subspecies × body mass 3.88 0.06 7.95 0.008 
Temperature × body mass 0.08 0.78 8.23 0.007 
Subspecies × Temperature × body mass 0.03 0.86 1.19 0.28 
 Log10 shell mass
 
Log10 hepatopancreas mass
 
F1, 39 P F1, 39 P 
Subspecies 2.17 0.15 13.66 0.0007 
Temperature 8.78 0.005 11.27 0.002 
body mass 241.69 0.000001 243.89 0.000001 
Subspecies × Temperature 0.22 0.64 4.24 0.046 
Subspecies × body mass 3.88 0.06 7.95 0.008 
Temperature × body mass 0.08 0.78 8.23 0.007 
Subspecies × Temperature × body mass 0.03 0.86 1.19 0.28 
Figure 2.

A. At a smaller body mass, the snails of Cornu aspersum aspersum had heavier hepatopancreas than C. a. maximum. At a larger body mass, C. a. aspersum had lighter hepatopancreas than C. a. maximum. B. Generally, snails of C. a. aspersum had faster metabolism than snails of C. a. maximum. Lines display the relationships estimated from a statistical model. Arrows labelled S, M and L indicate reference body masses, used to calculate differences in the mean hepatopancreas mass and metabolic rate between study groups.

Figure 2.

A. At a smaller body mass, the snails of Cornu aspersum aspersum had heavier hepatopancreas than C. a. maximum. At a larger body mass, C. a. aspersum had lighter hepatopancreas than C. a. maximum. B. Generally, snails of C. a. aspersum had faster metabolism than snails of C. a. maximum. Lines display the relationships estimated from a statistical model. Arrows labelled S, M and L indicate reference body masses, used to calculate differences in the mean hepatopancreas mass and metabolic rate between study groups.

Histological characteristics and cell size

The foot integument of snails was composed of transitional epithelium, but for cell size measurements only cuboidal cells were chosen and these were from areas where they formed a single layer (Fig. 3A, B). Nuclei of these cells had granulated chromatin and were oval in shape. In the subepithelial matrix, we identified mucus glands, connective tissue and muscle cells (Fig. 3A). The muscle cells occurred individually or formed bunches (Fig. 3C, D). Each muscle cell was surrounded by endomysium (Fig. 3D). In the hepatopancreas digestive cells were the most numerous cell type in the digestive-gland epithelium (Fig. 3E, F). Their shape varied from ovoid to, more frequently, columnar. The cytoplasm of these cells was highly vacuolated (Fig. 3F). The nuclei (typically one or, infrequently, two per cell) were round and they were located in the basal or middle region of the cells. Nuclei had from one to three nucleoli.

Figure 3.

A.Cornu aspersum maximum reared at 15 °C, fold of foot integument. B. C. a. maximum, 15 °C, external epithelium of foot integument and subepithelial connective tissue. C. C. a. maximum, 20 °C, muscle tissue. D. C. a. maximum, 20 °C, muscle tissue. E. C. a. maximum, 20 °C, hepatopancreas follicles. F. C. a. maximum, 20 °C, hepatopancreas cells. Abbreviations: CT, connective tissue; D, cytoplasm of digestive cells; E, foot epithelium; EM, endomysium; HP, hepatopancreas follicles; HS, system of haemocoelic sinuses (empty spaces); M, longitudinal and cross sections through muscle cells; MG, mucus gland; N, nucleus; Nu, nucleolus. Scale bars: A, C, E = 50 µm; B, D, F = 10 µm;

Figure 3.

A.Cornu aspersum maximum reared at 15 °C, fold of foot integument. B. C. a. maximum, 15 °C, external epithelium of foot integument and subepithelial connective tissue. C. C. a. maximum, 20 °C, muscle tissue. D. C. a. maximum, 20 °C, muscle tissue. E. C. a. maximum, 20 °C, hepatopancreas follicles. F. C. a. maximum, 20 °C, hepatopancreas cells. Abbreviations: CT, connective tissue; D, cytoplasm of digestive cells; E, foot epithelium; EM, endomysium; HP, hepatopancreas follicles; HS, system of haemocoelic sinuses (empty spaces); M, longitudinal and cross sections through muscle cells; MG, mucus gland; N, nucleus; Nu, nucleolus. Scale bars: A, C, E = 50 µm; B, D, F = 10 µm;

Table 3 summarizes the results of the PCA on the cell size in the three tissues. The size of muscle cells formed the first principal component (PC1), while the size of hepatopancreas cells formed the second principle component (PC2). The size of the epithelial cells contributed almost equally to PC1 and PC2, correlating positively with the size of the muscle cells (according to PC1) and negatively with the size of the hepatopancreas cells (according to PC2).

Table 3.

Loadings of three cell types in the principal component analysis of cell size. Scores for PC1 and PC2 were used as integrated measures of cell size.

 PC1 PC2 
Log10 muscle cell cross-section area 0.85 0.18 
Log10 muscle cell longitudinal-section diameter 0.79 0.30 
Log10 hepatopancreas cell cross-section area −0.12 0.82 
epithelial cell width 0.49 −0.60 
Explained variance % 40 29 
 PC1 PC2 
Log10 muscle cell cross-section area 0.85 0.18 
Log10 muscle cell longitudinal-section diameter 0.79 0.30 
Log10 hepatopancreas cell cross-section area −0.12 0.82 
epithelial cell width 0.49 −0.60 
Explained variance % 40 29 

The scores of PC1 were larger in CAA snails when compared with CAM snails, and in cold snails when compared to warm snails (Table 1, Fig. 4A, B). This indicates that CAA snails and the cold snails were characterized by large cells in the foot muscle and the foot epithelium. The scores of PC2 did not differ between CAA and CAM snails, but they were larger in the warm snails than in the cold snails (Table 1, Fig. 4A, B). This indicates that the taxa had similarly sized hepatopancreatic cells, but that the warm snails developed larger hepatopancreatic cells than the cold snails. As PC2 was formed by hepatopancreatic and epithelial cells, which were negatively related in size (Table 3), larger scores of PC2 of the warm snails also indicate that epithelial cells were smaller in the warm snails.

Figure 4.

A. Muscle cells and epithelial cells in the foot of adult snails were larger in Cornu aspersum aspersum than in C. a. maximum. The size of these cell types decreased in snails reared in a warm environment. B. The two snail subspecies had hepatopancreatic cells of the same size, but these cells were larger in warm-reared snails. Data are means ± 95% CI, estimated from a statistical model that analysed scores of principal components. Two principal components (PC1 and PC2) were formed by cell size in three types of tissue. Arrows indicate the sign and value of loadings of each cell type on a principal component (see Table 3 for details).

Figure 4.

A. Muscle cells and epithelial cells in the foot of adult snails were larger in Cornu aspersum aspersum than in C. a. maximum. The size of these cell types decreased in snails reared in a warm environment. B. The two snail subspecies had hepatopancreatic cells of the same size, but these cells were larger in warm-reared snails. Data are means ± 95% CI, estimated from a statistical model that analysed scores of principal components. Two principal components (PC1 and PC2) were formed by cell size in three types of tissue. Arrows indicate the sign and value of loadings of each cell type on a principal component (see Table 3 for details).

Our GLM of the karyoplasmic ratio in the hepatopancreas (Table 1) showed that this ratio did not differ among the taxa and the thermal treatments. This indicates that in hepatopancreatic cells the size of the nucleus changed in proportion to the size of the cells.

Metabolic rate

The rate of CO2 production was generally higher in the warm snails than in the cold snails (Table 4, Fig. 2B). This rate scaled allometrically with body mass in all experimental groups. A significant interaction among subspecies, thermal treatment and body mass indicates that this scaling was not uniform in all groups (Table 4). A mass-scaling exponent for the metabolic rate was equal to 0.84 in cold CAM, 0.88 in warm CAM, 0.89 in warm CAA and 0.95 in cold CAA. In warm snails, the rate of CO2 production was from 8 (small body mass) to 15% (large body mass) higher in CAA than CAM (Fig. 2B: arrows S and L). In cold snails, the rate of CO2 production at a small body mass (Fig. 2B: arrow S) was on average 5% lower in CAA than CAM. At larger body masses, CAA had either 16% (Fig. 2B: arrow M) or 41% (Fig. 2B: arrow L) higher rate of CO2 production compared with CAM.

Table 4.

Statistics for GLM of the rate of CO2 production (log10µlh−1) measured in snails age 136–150 d. Fixed factors: snail subspecies (Cornu aspersum aspersum and C. a. maximum) and developmental temperature (15 or 20 °C); log10 body mass is a covariate.

 F1, 112 P 
Subspecies 15.96 0.0001 
Temperature 43.28 0.000001 
body mass 910.80 0.000001 
Subspecies × Temperature 3.52 0.06 
Subspecies × body mass 0.24 0.63 
Temperature × body mass 4.67 0.03 
Subspecies × Temperature × body mass 10.12 0.002 
 F1, 112 P 
Subspecies 15.96 0.0001 
Temperature 43.28 0.000001 
body mass 910.80 0.000001 
Subspecies × Temperature 3.52 0.06 
Subspecies × body mass 0.24 0.63 
Temperature × body mass 4.67 0.03 
Subspecies × Temperature × body mass 10.12 0.002 

DISCUSSION

At the end of our common garden experiment, snails showed signs of maturation and warm snails were larger than cold snails. Although we did not examine size differences at maturation, our results suggest that the snails of Cornu aspersum do not follow the temperature-size rule (Atkinson, 1994; Forster et al., 2012; Klok & Harrison, 2013) and that a warm environment increases their growth rate as well as adult size. Using histological techniques, we found that snails grown in the warmer environment produced smaller muscle and epithelial cells, but larger hepatopancreatic cells. These results indicate that the temperature-size rule for body size cannot be explained as a simple consequence of thermal sensitivity of cell size, as has been suggested before (van Voorhies, 1996).

Focusing on the patterns between the two subspecies, we found that C. aspersum aspersum grew slowly and attained a smaller adult mass than C. aspersum maximum. What is more, the body of CAA consisted of larger muscular and epithelial cells when compared with CAM, regardless of temperature, and both subspecies had a similar size of hepatopancreatic cells. This suggests that the divergence of C. aspersum into subspecies aspersum and maximum during the Pliocene involved genetic differentiation in life history and, consequently, in adult mass, but that this size divergence was not a simple scaling effect of changes in cell size. The primary geographical distributions of CAA and CAM remain unclear (Guiller et al., 2001) but, given their current distribution, we conclude that the body size differences between the snails do not comply with Bergmann's rule, which predicts larger body size in species or populations from higher latitudes (Ashton, 2002; Angilletta et al., 2004). If our findings generalize to other species of ectotherms, the origin of latitudinal clines in body size might not be a direct effect of latitudinal differentiation of cell sizes. Certainly, this hypothesis needs further testing.

Our histological results have important implications. First, body size and cell size appear to change in a more complex way than is usually assumed, so that a decrease in cell size does not mechanically translate into a decrease in body size. In fact, the cell size of Drosophila melanogaster has been shown to change in part independently of body size in response to developmental conditions (Czarnoleski et al., 2013, 2015) and mice divergently selected for basal metabolic rate evolve cell size without an alteration in body mass (Maciak et al., 2014). In eutelic organisms, whose body volume consists of a fixed number of cells, body size and cell size cannot respond freely to opposing selective pressures. However, our evidence suggests that in noneutelic organisms the effects of selection pressure on cell size are not constrained by selective pressure on body size. Second, our results indicate that a change in the size of one cell type does not necessarily involve changes in the size of other cells. Earlier studies suggested that evolutionary changes in the size of different cell types occurred in a coordinated manner in invertebrates (Stevenson, Hill & Bryant, 1995), vertebrates (Kozlowski et al., 2010) and plants (Brodribb, Jordan & Carpenter, 2013). Additionally, concerted changes in cell sizes were induced in drosophilids by developmental conditions (Azevedo, French & Partridge, 2002; Heinrich et al., 2011). Nevertheless, irregularities in the coordinated evolution of cell size were reported by Kozlowski et al. (2010) and Maciak et al. (2014).

The thermal plasticity of cell size revealed by our experiment is only partially consistent with the concept of optimal cell size (Szarski, 1983; Kozlowski et al., 2003). Although warm snails had small cells in the epithelium and muscles, they had large cells in the hepatopancreas. In warm environments, demands for resources are high compared with supplies of oxygen (Woods, 1999; Atkinson et al., 2006; Verberk et al., 2011). Small cells should help to increase the physiological performance at high temperatures by providing sufficient surface area to transport oxygen and nutrients, and short distances within cytoplasm for transport of resources and macromolecules (Atkinson, et al., 2006; Czarnoleski, et al., 2013). In agreement, many ectotherms—including bacteria (Sjostedt, Hagstrom & Zweifel, 2012), protists (Butler & Rogerson, 1996; Atkinson, Ciotti & Montagnes, 2003), rotifers (Stelzer, 2002), planarians (Romero & Baguna, 1991), nematodes (van Voorhies, 1996), flies (Partridge et al., 1994; Blanckenhorn & Llaurens, 2005), fish (van Voorhies, 1996) and lizards (Goodman & Heah, 2010)—develop smaller cells in constant warm environments. Moreover, thermal fluctuations during development cause flies to develop smaller cells, which suggests the role of small cells in coping with even brief periods at elevated temperatures (Czarnoleski et al., 2013). This explanation, however, cannot account for the pattern observed in our experiment because all tissues of snails experienced the same temperatures for the same duration, yet one cell type became smaller whereas the other cells became larger in a warm environment. Therefore, we must consider the tissue-specific functions of cells. According to Maciak et al. (2014), increased physiological performance can require a reduction in cell size in tissues whose functions are dictated by surface-to-volume ratios, but an increase in the size of cells and their nuclei in tissues with high biochemical activity. Because the intensity of transcription and translation depends on the density of molecules in the nucleus and cytoplasm (Zimmerman, 1993; Ellis, 2001), large cells should be less prone to a dumping effect of molecular crowding than small cells. Our results are consistent with this view. First, warm snails had smaller cells in the tissues with mechanical functions (epithelium and locomotory muscle in the foot), but had larger cells in the anabolically-active tissue (hepatopancreas). Second, the karyoplasmic ratio of the hepatopancreatic cells was constant, indicating that changes in the size of hepatopancreatic cells were followed by a proportional change in the size of the nucleus. Mechanisms that control the karyoplasmic ratio and the functional significance of an invariant karyoplasmic ratio are still a major puzzle in cell biology (Cavalier-Smith, 2005; Cohen-Fix, 2010), but we suggest that this invariance might indicate the tuning of translational and transcriptional activity of cells to the physiological performance of an organism.

Despite the likely connection between cell size and evolutionary fitness, the ecological relevance of this link remains unexplored. One of potential ecological consequences of cell size might be the capacity to colonize thermally novel environments. We found that the snails of subspecies CAA, which expanded from the warm subtropics to cool temperate regions (Guiller, et al., 2012), consist of larger cells in at least two tissues types compared with the noninvasive CAM snails. All else being equal, a volume of tissue that consists of larger cells must utilize per volume less resources for the turnover of cell membrane components and the generation of ionic gradients across cell membranes (Szarski, 1983; Kozlowski et al., 2003; Czarnoleski et al., 2015). Therefore, it is tempting to use the theory of optimal cell size to speculate that larger cells of CAA reduce the cost associated with tissue maintenance, which ultimately helps CAA to deal with cooler environments. Obviously, before drawing firm conclusion, it should be determined how general these findings may be. Apparently, snails of C. aspersum that originate from different latitudes differ in their thermal optima for performance (Gaitan-Espitia et al., 2013), but we are not aware of any study that has attempted to link such differences in performance to variation in cell size. Previous works have demonstrated an inverse relation between cell size and metabolic rates in geckos (Starostova et al., 2009) and fish (Maciak et al., 2011), but we did not find evidence that cell size accounts for the observed differences in metabolic rates between CAA and CAM—the snails of CAA, which were characterized by slow growth and large cells, had higher mass-specific metabolic rates than the fast-growing CAM. Most likely, the metabolic differences between CAA and CAM manifest a physiological trade-off between growth and other metabolically-demanding functions. Although growth and shell production can compete for resources in C. aspersum (Czarnoleski et al., 2008), we found no evidence of this trade-off in the studied snails—both subspecies produced shells of similar mass and, despite their fast growth, warm snails produced heavier shells than cold snails. Yet, we found an indication that CAA had a larger hepatopancreas, at least at smaller body masses. Similarly to the liver of vertebrates, the hepatopancreas of invertebrates performs multiple service functions that require substantial amounts of ATP (Rolfe & Brown, 1997; Pakay et al., 2002). Given that a large mass of tissue performs more work than a small mass of tissue, CAA snails were likely using more energy in their hepatopancreas, draining resources from growth, which ultimately raised the rate of metabolism at an organismal level.

The key message of this study is that we must consider the physiological effects of cell size to understand better the relationships between organismal performance, evolutionary fitness, body size, cell size and environmental conditions. If the ecological patterns described by Bergmann's rule and the temperature-size rule prove to prevail in nature, then the subspecies of C. aspersum would be one of the exceptions. By showing that warm snails grow larger body mass but smaller cells, we have demonstrated that these two ecological rules are not a simple consequence of developmental constraints on body size imposed by the thermal sensitivity of cell size. Nevertheless, modelling of resource allocation in ectotherms demonstrates that the physiological consequences of a change in cell size should not be ignored. These consequences can bring about adaptive changes in resource allocation, which can produce patterns in body size that comply with Bergmann's rule and the temperature-size rule (Kozlowski, Czarnoleski & Danko, 2004). Certainly, this emerging concept needs further efforts to evaluate empirically the physiological consequences of cell size and its effects on life-history strategy.

ACKNOWLEDGEMENTS

The research was supported by a Maestro grant from the Polish National Science Centre (2011/02A/NZ8/00064). A.M.L. was supported by the program ‘DS for Development of Young Researchers and PhD Students' of the Faculty of Biology and Earth Sciences, Jagiellonian University (K/DSC/000171). Data on snail growth were collected and used in the MSc theses of Łukasz Hołda and Sabina Papuga. Katarzyna Pawlik measured the epithelial cells, Anna Sikorska made the gonad slides and Ulf Bauchinger commented on earlier versions of the manuscript. We thank two anonymous reviewers for their constructive suggestions.

REFERENCES

ANGILLETTA
M.J.
,
NIEWIAROWSKI
P.H.
,
DUNHAM
A.E.
,
LEACHE
A.D.
,
PORTER
W.P.
2004
.
Bergmann's clines in ectotherms: illustrating a life-history perspective with sceloporine lizards
.
American Naturalist
 ,
164
:
E168
E183
.
ARENDT
J.
2007
.
Ecological correlates of body size in relation to cell size and number: patterns in flies, fish, fruits, and foliage
.
Biological Reviews
 ,
82
:
241
256
.
ASHTON
K.G.
2002
.
Do amphibians follow Bergmann's rule?
Canadian Journal of Zoology
 ,
80
:
708
716
.
ATKINSON
D.
1994
.
Temperature and organism size: a biological law for ectotherms
.
Advances in Ecological Research
 ,
25
:
1
58
.
ATKINSON
D.
,
CIOTTI
B.J.
,
MONTAGNES
D.J.S.
2003
.
Protists decrease in size linearly with temperature: ca. 2.5% degrees C-1
.
Proceedings of the Royal Society B
 ,
270
:
2605
2611
.
ATKINSON
D.
,
MORLEY
S.A.
,
HUGHES
R.N.
2006
.
From cells to colonies: at what levels of body organization does the ‘temperature-size rule’ apply?
Evolution & Development
 ,
8
:
202
214
.
AZEVEDO
R.B.R.
,
FRENCH
V.
,
PARTRIDGE
L.
2002
.
Temperature modulates epidermal cell size in Drosophila melanogaster
.
Journal of Insect Physiology
 ,
48
:
231
237
.
BLANCKENHORN
W.U.
,
LLAURENS
V.
2005
.
Effects of temperature on cell size and number in the yellow dung fly Scathophaga stercoraria
.
Journal of Thermal Biology
 ,
30
:
213
219
.
BRODRIBB
T.J.
,
JORDAN
G.J.
,
CARPENTER
R.J.
2013
.
Unified changes in cell size permit coordinated leaf evolution
.
New Phytologist
 ,
199
:
559
570
.
BUTLER
H.
,
ROGERSON
A.
1996
.
Growth potential, production efficiency and annual production of marine benthic naked amoebae (gymnamoebae) inhabiting sediments of the Clyde Sea area, Scotland
.
Aquatic Microbial Ecology
 ,
10
:
123
129
.
CAVALIER-SMITH
T.
2005
.
Economy, speed and size matter: Evolutionary forces driving nuclear genome miniaturization and expansion
.
Annals of Botany
 ,
95
:
147
175
.
COHEN-FIX
O.
2010
.
CELL BIOLOGY Import and nuclear size
.
Nature
 ,
468
:
513
516
.
CZARNOLESKI
M.
,
COOPER
B.S.
,
KIERAT
J.
,
ANGILLETTA
M.J.
2013
.
Flies developed small bodies and small cells in warm and in thermally fluctuating environments
.
Journal of Experimental Biology
 ,
216
:
2896
2901
.
CZARNOLESKI
M.
,
DRAGOSZ-KLUSKA
D.
,
ANGILLETTA
M.J.
2015
.
Flies developed smaller cells when temperature fluctuated more frequently
.
Journal of Thermal Biology
 ,
in press
.
CZARNOLESKI
M.
,
KOZLOWSKI
J.
,
DUMIOT
G.
,
BONNET
J.C.
,
MALLARD
J.
,
DUPONT-NIVET
M.
2008
.
Scaling of metabolism in Helix aspersa snails: changes through ontogeny and response to selection for increased size
.
Journal of Experimental Biology
 ,
211
:
391
399
.
CZARNOLESKI
M.
,
KOZŁOWSKI
J.
,
STANCZYKOWSKA
A.
,
LEWANDOWSKI
K.
2003
.
Optimal resource allocation explains growth curve diversity in zebra mussels
.
Evolutionary Ecology Research
 ,
5
:
571
587
.
ELLIS
R.J.
2001
.
Macromolecular crowding: obvious but underappreciated
.
Trends in Biochemical Sciences
 ,
26
:
597
604
.
FORSTER
J.
,
HIRST
A.G.
,
ATKINSON
D.
2012
.
Warming-induced reductions in body size are greater in aquatic than terrestrial species
.
Proceedings of the National Academy of Sciences of the USA
 ,
109
:
19310
19314
.
GAITAN-ESPITIA
J.D.
,
ARIAS
M.B.
,
LARDIES
M.A.
,
NESPOLO
R.F.
2013
.
Variation in thermal sensitivity and thermal tolerances in an invasive species across a climatic gradient: lessons from the land snail Cornu aspersum
.
PLoS One
 ,
8
8
:
e70662
.
GOODMAN
R.M.
,
HEAH
T.P.
2010
.
Temperature-induced plasticity at cellular and organismal levels in the lizard Anolis carolinensis
.
Integrative Zoology
 ,
5
:
208
217
.
GUILLER
A.
,
BELLIDO
A.
,
COUTELLE
A.
,
MADEC
L.
2006
.
Spatial genetic pattern in the land mollusc Helix aspersa inferred from a ‘centre-based clustering’ procedure
.
Genetical Research
 ,
88
:
27
44
.
GUILLER
A.
,
COUTELLEC-VRETO
M.A.
,
MADEC
L.
,
DEUNFF
J.
2001
.
Evolutionary history of the land snail Helix aspersa in the Western Mediterranean: preliminary results inferred from mitochondrial DNA sequences
.
Molecular Ecology
 ,
10
:
81
87
.
GUILLER
A.
,
MARTIN
M.C.
,
HIRAUX
C.
,
MADEC
L.
2012
.
Tracing the invasion of the Mediterranean land snail Cornu aspersum aspersum becoming an agricultural and garden pest in areas recently introduced
.
PLoS One
 ,
7
12
:
e49674
.
HEINRICH
E.
,
FARZIN
M.
,
KLOK
C.
,
HARRISON
J.
2011
.
The effect of developmental stage on the sensitivity of cell and body size to hypoxia in Drosophila melanogaster
.
Journal of Experimental Biology
 ,
214
:
1419
1427
.
KLOK
C.J.
,
HARRISON
J.F.
2013
.
The temperature size rule in arthropods: independent of macro-environmental variables but size dependent
.
Integrative and Comparative Biology
 ,
53
:
557
570
.
KOZŁOWSKI
J.
1992
.
Optimal allocation of resources to growth and reproduction: implications for age and size at maturity
.
Trends in Ecology & Evolution
 ,
7
:
15
19
.
KOZLOWSKI
J.
,
CZARNOLESKI
M.
,
DANKO
M.
2004
.
Can optimal resource allocation models explain why ectotherms grow larger in cold?
Integrative and Comparative Biology
 ,
44
:
480
493
.
KOZLOWSKI
J.
,
CZARNOLESKI
M.
,
FRANCOIS-KRASSOWSKA
A.
,
MACIAK
S.
,
PIS
T.
2010
.
Cell size is positively correlated between different tissues in passerine birds and amphibians, but not necessarily in mammals
.
Biology Letters
 .
KOZLOWSKI
J.
,
KONARZEWSKI
M.
,
GAWELCZYK
A.T.
2003
.
Cell size as a link between noncoding DNA and metabolic rate scaling
.
Proceedings of the National Academy of Sciences of the USA
 ,
100
:
14080
14085
.
ŁYSAK
A.
,
MACH-PALUSZKIEWICZ
Z.
,
LIGASZEWSKI
M.
2001
.
Fermowy wychów ślimaków jadalnych. Materiały szkoleniowe na konferencję hodowców ślimaków
 .
Krakow, Poland, Instytut Zootechniki w Balicach koło Krakowa
.
MACIAK
S.
,
BONDA-OSTASZEWSKA
E.
,
CZARNOLESKI
M.
,
KONARZEWSKI
M.
,
KOZLOWSKI
J.
2014
.
Mice divergently selected for high and low basal metabolic rates evolved different cell size and organ mass
.
Journal of Evolutionary Biology
 ,
27
:
478
487
.
MACIAK
S.
,
JANKO
K.
,
KOTUSZ
J.
,
CHOLEVA
L.
,
BORON
A.
,
JUCHNO
D.
,
KUJAWA
R.
,
KOZLOWSKI
J.
,
KONARZEWSKI
M.
2011
.
Standard metabolic rate (SMR) is inversely related to erythrocyte and genome size in allopolyploid fish of the Cobitis taenia hybrid complex
.
Functional Ecology
 ,
25
:
1072
1078
.
PAKAY
J.L.
,
WITHERS
P.C.
,
HOBBS
A.A.
,
GUPPY
M.
2002
.
In vivo downregulation of protein synthesis in the snail Helix apersa during estivation
.
American Journal of Physiology—Regulatory Integrative and Comparative Physiology
 ,
283
:
R197
R204
.
PARTRIDGE
L.
,
BARRIE
B.
,
FOWLER
K.
,
FRENCH
V.
1994
.
Evolution and development of body size and cell size in Drosophila melanogaster in response to temperature
.
Evolution
 ,
48
:
1269
1276
.
ROLFE
D.F.S.
,
BROWN
G.C.
1997
.
Cellular energy utilization and molecular origin of standard metabolic rate in mammals
.
Physiological Reviews
 ,
77
:
731
758
.
ROMERO
R.
,
BAGUNA
J.
1991
.
Quantitative cellular analysis of growth and reproduction in fresh-water planarians (Turbellaria, Tricladida). I. A cellular description of the intact organism
.
Invertebrate Reproduction & Development
 ,
19
:
157
165
.
SJOSTEDT
J.
,
HAGSTROM
A.
,
ZWEIFEL
U.L.
2012
.
Variation in cell volume and community composition of bacteria in response to temperature
.
Aquatic Microbial Ecology
 ,
66
:
237
246
.
STAROSTOVA
Z.
,
KUBICKA
L.
,
KONARZEWSKI
M.
,
KOZLOWSKI
J.
,
KRATOCHVIL
L.
2009
.
Cell size but not genome size affects scaling of metabolic rate in eyelid geckos
.
American Naturalist
 ,
174
:
E100
E105
.
STEARNS
S.C.
1992
.
The evolution of life histories
 .
Oxford University Press
,
Oxford
.
STELZER
C.P.
2002
.
Phenotypic plasticity of body size at different temperatures in a planktonic rotifer: mechanisms and adaptive significance
.
Functional Ecology
 ,
16
:
835
841
.
STEVENSON
R.D.
,
HILL
M.F.
,
BRYANT
P.J.
1995
.
Organ and cell allometry in Hawaiian Drosophila: how to make a big fly
.
Proceeding of the Royal Society B
 ,
259
:
105
110
.
SZARSKI
H.
1983
.
Cell size and the concept of wasteful and frugal evolutionary strategies
.
Journal of Theoretical Biology
 ,
105
:
201
209
.
VAN VOORHIES
W.A.
1996
.
Bergmann size clines: a simple explanation for their occurrence in ectotherms
.
Evolution
 ,
50
:
1259
1264
.
VERBERK
W.
,
BILTON
D.T.
,
CALOSI
P.
,
SPICER
J.I.
2011
.
Oxygen supply in aquatic ectotherms: partial pressure and solubility together explain biodiversity and size patterns
.
Ecology
 ,
92
:
1565
1572
.
WALCZYŃSKA
A.
,
LABECKA
A.M.
,
SOBCZYK
M.
,
CZARNOLESKI
M.
,
KOZLOWSKI
J.
2015
.
The temperature-size rule in Lecane inermis (Rotifera) is adaptive and driven by nuclei size adjustment to temperature and oxygen concentrations
.
Journal of Thermal Biology
 .
WOODS
H.A.
1999
.
Egg-mass size and cell size: effects of temperature on oxygen distribution
.
American Zoologist
 ,
39
:
244
252
.
ZIMMERMAN
S.B.
1993
.
Macromolecular crowding effects on macromolecular interactions—some implications for genome structure and function
.
Biochimica et Biophysica Acta
 ,
1216
:
175
185
.
ZWAAN
B.J.
,
AZEVEDO
R.B.R.
,
JAMES
A.C.
,
VAN'T LAND
J.
,
PARTRIDGE
L.
2000
.
Cellular basis of wing size variation in Drosophila melanogaster: a comparison of latitudinal clines on two continents
.
Heredity
 ,
84
:
338
347
.