**3.1 Protein variation between individual** *D. magna* **of distinct genotypes**

Our first goal was to demonstrate that differences in protein expression could be detected between individual *Daphnia* of distinct genotypes. Individual *D. magna* from Iinb1 and Xinb3 genotypes were isolated, proteins extracted and analyzed in quadruplicate by 2-DE as

cycles, each cycle consisting of 10 seconds at 35,000 psi followed by rapid depressurization and held for 2 seconds at atmospheric pressure. Following PCT, each PULSE Tube was coupled to an Ultrafree-CL centrifugal filtration device with a 5-micron pore size (Millipore Corporation, Danvers, MA, USA) and evacuated by centrifugation for 1 minute at 1000 RCF. The PULSE Tube was removed and centrifugation continued for 4 minutes at 4000 RCF,

Samples were transferred to ULTRA-4 ultrafiltration devices with 10 kDa molecular weight cut-off (Millipore Corporation, Danvers, MA, USA). Proteins larger than 10 kDa are retained in the ultrafiltration device. Centrifugation assisted the ultrafiltration, and the samples were exchanged with fresh UTC until the final DTT concentration was 10 mM. Reduction and alkylation of the samples were performed directly in the ultrafiltration devices using 5 mM tributylphosphine and 50 mM acrylamide as described (Smejkal et al., 2006a). The alkylation reaction was terminated by resuming centrifugation and ultrafiltrative exchange. Bradford Reagent (Sigma-Aldrich Chemicals, St. Louis, MO, USA) was used to measure the protein

Two-hundred uL of each sample was placed onto individual wells in IPG rehydration trays from Proteome Systems (Woburn, MA, USA). Bio-Rad ReadyStrip IPG strips with a pH range of 3-10, 4-7, or 7-10 (Hercules, CA, USA) were placed onto each sample, and the tray was placed into a humidifying Ziploc bag. Rehydration occurred over six hours until all the sample was visibly absorbed by the strip. At the termination of rehydration, strips were placed into isoelectric focusing trays and ran at 10,000 volts (maximum voltage) for 110,000 accumulative volt-hours. Strips were equilibrated twice for 10 minutes in 375 mM Tris-HCl containing 2.5% SDS, 3M urea, 0.01% and phenol red, then placed onto Criterion Tris-HCl 8- 16% IPG+1 gels (Bio-Rad Laboratories, Hercules, CA) and ran at 120 V and 60 mA/gel. 2-DE

The 24-bit images were analyzed using PDQuest™ software (Bio-Rad, v.7.1). Background was subtracted, and protein spot density peaks were detected and counted. A reference pattern was constructed from one of the individual gels to which each of the gels in the matchset was matched. Numerous proteins that were uniformly expressed in all patterns were used as landmarks to facilitate rapid gel matching. After matching, the total spot count

Our first goal was to demonstrate that differences in protein expression could be detected between individual *Daphnia* of distinct genotypes. Individual *D. magna* from Iinb1 and Xinb3 genotypes were isolated, proteins extracted and analyzed in quadruplicate by 2-DE as

**3.1 Protein variation between individual** *D. magna* **of distinct genotypes** 

followed by the removal of the mineral oil.

concentration in each sample.

**2.6 Digital image analyses** 

was determined for each gel.

**3. Results and discussion** 

**2.5 IEF and 2-DE** 

**2.4 Reduction, alkylation, and ultrafiltration** 

gels were ran for all IPGs, only 4-7 gels are shown.

describe above. Silver staining detected an average of 687 + 11 protein spots from the Xinb3 gels and 692 + 14 protein spots from the Iinb1 gels (figure 1). After normalization of the gel images based on total intensity, 679 spots were matched between the two gel images. One Hundred thirty six spots showed a two-fold or greater difference in spot intensity. Seventynine of these were more abundant in Xinb3 and 57 were more abundant in the Iinb1. To illustrate these differences, the silver stained gels were digitally colored. The Iinb1 gels were red and the Xinb3 gels were green. The gels were then superimposed. (Figure 2).

Fig. 1. Representative silver-stained 2D gels of individual *Daphnia magna* with distinct genotypes. Quadruplicate gels revealed 692 + 14 spots in Iinb1 clone (left) and 687 + 11 spots in Xinb3 clone (right).

Fig. 2. Digitally enhanced, superimposed, silver-stained 2D gels of individual *Daphnia magna* with distinct genotypes. Red indicates spots unique to Iinb1, green indicates spots unique to Xinb3, yellow indicate spots shared by both genotypes.

Two-Dimensional Gel Electrophoresis Reveals

**3.3 Extending approach to single** *D. pulex*

comparison.

Differential Protein Expression Between Individual *Daphnia* 321

As *D. magna* is the largest of the *Daphnia* genus, we used PCT to extract protein from single *D. pulex* to demonstrate that our technique would be feasible with smaller individuals. Even when using the much smaller *Daphnia pulex,* we were able to detect approximately 900 spots from a single individual. It is reasonable to expect that differences in 2-DE between single *D. pulex* with phenotypic differences would also be detectable. As an indication of reproducibility of our method, we ran duplicate 2-DE of PCT- extracted proteins from 1, 2, 3, 4 and 5 *Daphnia pulex*. Figure 4 shows representative gels of 1, 2 or 3 *Daphnia pulex*. Figure 5 shows the number of protein spots detected and the standard deviation for *Daphnia pulex*  gels of 1, 2, 3, 4 and 5 individuals. The low standard deviation indicates that PCT and 2-DE is an efficient and highly reproducible method of sample preparation and protein

Fig. 4. Representative, silver stained 2D gels of 1, 2 & 3 *Daphnia pulex.* 

#### **3.2 Protein variation between individual** *D. magna* **of distinct phenotypes**

We were also able to detect differences between individual *Daphnia* with distinct phenotypic differences. Individual *D. magna*, with and without ephippia, were isolated, proteins extracted and analyzed by 2-DE as described above. Silver staining detected an average of 524.5 ± 7.8 protein spots in 2D gels produced from single *D. magna,* with and without ephippia (figure 3). After normalization of the gel images based on total intensity, 386 spots were matched between the two gel images. Eighty-four spots showed a three-fold or greater difference in spot intensity. Fifty-five of these were more abundant in the parthenogenic (no ephippia) animal, while 29 were more abundant in the sexual animal. In addition, eleven protein spots were unique to the parthenogenic phenotype, while 49 protein spots were unique to the sexual phenotype. This demonstrates the feasibility of 2-DE and image analysis for the differentiation of *Daphnia* phenotypes isolated in the field as indicators of environmental variables. Other studies with parthenogenic and sexual *Daphnia carinata* were able to identify several proteins that were differentially expressed between the two phenotypes by 2-DE; however, 100's of animals were used (Zhang et al., 2006). It is interesting to note that using single animals, we discovered similar patterns of up-regulation in the parthenogenic individual in comparison to the sexual phenotype. While Zhang et al.'s goal was only to gain insight into the genes involved in the switch to sexual reproduction, our method, using single *Daphnia magna* with and without ephippia, allows the sampling of these candidate genes within a population.

Fig. 3. Silver stained 2D gels of individual *Daphnia magna* with unique phenotypes.

#### **3.3 Extending approach to single** *D. pulex*

320 Gel Electrophoresis – Advanced Techniques

We were also able to detect differences between individual *Daphnia* with distinct phenotypic differences. Individual *D. magna*, with and without ephippia, were isolated, proteins extracted and analyzed by 2-DE as described above. Silver staining detected an average of 524.5 ± 7.8 protein spots in 2D gels produced from single *D. magna,* with and without ephippia (figure 3). After normalization of the gel images based on total intensity, 386 spots were matched between the two gel images. Eighty-four spots showed a three-fold or greater difference in spot intensity. Fifty-five of these were more abundant in the parthenogenic (no ephippia) animal, while 29 were more abundant in the sexual animal. In addition, eleven protein spots were unique to the parthenogenic phenotype, while 49 protein spots were unique to the sexual phenotype. This demonstrates the feasibility of 2-DE and image analysis for the differentiation of *Daphnia* phenotypes isolated in the field as indicators of environmental variables. Other studies with parthenogenic and sexual *Daphnia carinata* were able to identify several proteins that were differentially expressed between the two phenotypes by 2-DE; however, 100's of animals were used (Zhang et al., 2006). It is interesting to note that using single animals, we discovered similar patterns of up-regulation in the parthenogenic individual in comparison to the sexual phenotype. While Zhang et al.'s goal was only to gain insight into the genes involved in the switch to sexual reproduction, our method, using single *Daphnia magna* with and without ephippia, allows the sampling of

**3.2 Protein variation between individual** *D. magna* **of distinct phenotypes**

Fig. 3. Silver stained 2D gels of individual *Daphnia magna* with unique phenotypes.

these candidate genes within a population.

As *D. magna* is the largest of the *Daphnia* genus, we used PCT to extract protein from single *D. pulex* to demonstrate that our technique would be feasible with smaller individuals. Even when using the much smaller *Daphnia pulex,* we were able to detect approximately 900 spots from a single individual. It is reasonable to expect that differences in 2-DE between single *D. pulex* with phenotypic differences would also be detectable. As an indication of reproducibility of our method, we ran duplicate 2-DE of PCT- extracted proteins from 1, 2, 3, 4 and 5 *Daphnia pulex*. Figure 4 shows representative gels of 1, 2 or 3 *Daphnia pulex*. Figure 5 shows the number of protein spots detected and the standard deviation for *Daphnia pulex*  gels of 1, 2, 3, 4 and 5 individuals. The low standard deviation indicates that PCT and 2-DE is an efficient and highly reproducible method of sample preparation and protein comparison.

Fig. 4. Representative, silver stained 2D gels of 1, 2 & 3 *Daphnia pulex.* 

Two-Dimensional Gel Electrophoresis Reveals

relevant proteins.

Differential Protein Expression Between Individual *Daphnia* 323

"coenzyme transport and metabolism", "secondary metabolites, biosynthesis, transport and catabolism", "nuclear structure" and "chromatin structure and dynamics" (Figure 7). In general, KOG classes that are well represented in the *Daphnia* genome are also well represented in the top 2000 of most highly expressed genes. As 84% of the KOG classes are represented in the top 1000 most highly expressed genes, we feel that a single animal 2D gel of approximately 1000 spots should represent a diverse sampling of biologically

Fig. 6. Theoretical 2D gel of the top 2000 highly expressed *D. pulex* genes. The rectangle

 **theoretical of total observed coefficient pI range proteins** *a* **predicted proteins of variation**  3-4 17 0.9 - - 4-7 1027 51.4 904.0 ± 7.0 *b* 0.008 7-10 749 37.5 381.5 ± 26.5 *c* 0.069 > 10 207 10.4 - **total** 2000 - 1285.5 ± 33.5 0.026

Table 1. Distribution of 2000 most abundant *Daphnia* proteins predicted from genome

represents the separation range (MW and pI) of actual 2D gels (pH range 4 – 7).

 **number of percent number of** 

a Proteins predicted from sequence does not account for charge b Spots detected in 2D gels from single D. pulex organisms. c Spots detected in 2D gels of a multiple D. magna organisms.

(184.0 ± 2.6 mg or approximately 130 daphnids)

sequence

Fig. 5. Graph of detected protein spots from duplicate silver stained 2D gels of proteins extracted 1, 2, 3, 4 & 5 *Daphnia pulex.* The shape of the curve suggests that our method is an efficient and reproducible approach to protein extraction and the majority of the unique proteins are recovered from a single animal.

#### **3.4 Protein functional diversity**

To properly evaluate our method, it is necessary to understand the functional diversity of the proteins sampled. The proteins detected on any 2D gel will be biased towards the most abundant and the goal is to sample gene expression for proteins of diverse functions. To evaluate the likely diversity of proteins detectable and comparable by this method, we first generated a theoretical 2D gel for the 2000 most highly expressed genes in the *Daphnia pulex* genome. The correlation between mRNA expression levels and protein abundance is a debated topic; (for a brief review, see Greenbaum et al., 2003) however, recent studies have found a high correlation (Lu et al., 2007; Tuller et al., 2007). Using the recently completed draft of the *Daphnia pulex* genome, we found the top 2000 gene prediction models with the most Expressed Sequence Tag (EST) evidence using BLAST (Altschul et al., 1990). The theoretical pI and MW of these genes were calculated using the Compute pI/MW tool from ExPASy (Gasteiger et al., 2003) and the results graphed using Excel to create the theoretical 2D gel (Figure 6). Importantly, the pI range used on the actual gels (indicated by the rectangle) covers a significant portion of the predicted proteome.

To assay the functional diversity of these theoretical proteins, we utilized the 25 eukaryotic orthologous groups (KOGs) (Totusov et al., 2003) that were assigned to the *Daphnia pulex* genes as part of the genome sequence annotation project. Many *D. pulex* genes have no homology to any entries at NCBI; therefore, it is not surprising that of the approximately 30,000 predicted genes only 18,371 have been assigned to a KOG class (Colbourne et al., 2011) Of the 2000 most highly expressed genes, 298 had been assigned to a KOG class. Twenty four of the 25 KOG classes were represented, and only "coenzyme transport and metabolism" was absent. To understand the functional diversity that may be present on a single animal 2D gel of approximately 1000 spots, we looked at the KOG assignments of the 1000 most highly expressed genes. Ninety-seven have been assigned to a KOG class, with 21 of the 25 KOG classes being represented. Not represented were

Fig. 5. Graph of detected protein spots from duplicate silver stained 2D gels of proteins extracted 1, 2, 3, 4 & 5 *Daphnia pulex.* The shape of the curve suggests that our method is an efficient and reproducible approach to protein extraction and the majority of the unique

To properly evaluate our method, it is necessary to understand the functional diversity of the proteins sampled. The proteins detected on any 2D gel will be biased towards the most abundant and the goal is to sample gene expression for proteins of diverse functions. To evaluate the likely diversity of proteins detectable and comparable by this method, we first generated a theoretical 2D gel for the 2000 most highly expressed genes in the *Daphnia pulex* genome. The correlation between mRNA expression levels and protein abundance is a debated topic; (for a brief review, see Greenbaum et al., 2003) however, recent studies have found a high correlation (Lu et al., 2007; Tuller et al., 2007). Using the recently completed draft of the *Daphnia pulex* genome, we found the top 2000 gene prediction models with the most Expressed Sequence Tag (EST) evidence using BLAST (Altschul et al., 1990). The theoretical pI and MW of these genes were calculated using the Compute pI/MW tool from ExPASy (Gasteiger et al., 2003) and the results graphed using Excel to create the theoretical 2D gel (Figure 6). Importantly, the pI range used on the actual gels (indicated by the rectangle) covers a significant portion of the predicted

To assay the functional diversity of these theoretical proteins, we utilized the 25 eukaryotic orthologous groups (KOGs) (Totusov et al., 2003) that were assigned to the *Daphnia pulex* genes as part of the genome sequence annotation project. Many *D. pulex* genes have no homology to any entries at NCBI; therefore, it is not surprising that of the approximately 30,000 predicted genes only 18,371 have been assigned to a KOG class (Colbourne et al., 2011) Of the 2000 most highly expressed genes, 298 had been assigned to a KOG class. Twenty four of the 25 KOG classes were represented, and only "coenzyme transport and metabolism" was absent. To understand the functional diversity that may be present on a single animal 2D gel of approximately 1000 spots, we looked at the KOG assignments of the 1000 most highly expressed genes. Ninety-seven have been assigned to a KOG class, with 21 of the 25 KOG classes being represented. Not represented were

proteins are recovered from a single animal.

**3.4 Protein functional diversity** 

proteome.

"coenzyme transport and metabolism", "secondary metabolites, biosynthesis, transport and catabolism", "nuclear structure" and "chromatin structure and dynamics" (Figure 7). In general, KOG classes that are well represented in the *Daphnia* genome are also well represented in the top 2000 of most highly expressed genes. As 84% of the KOG classes are represented in the top 1000 most highly expressed genes, we feel that a single animal 2D gel of approximately 1000 spots should represent a diverse sampling of biologically relevant proteins.

Fig. 6. Theoretical 2D gel of the top 2000 highly expressed *D. pulex* genes. The rectangle represents the separation range (MW and pI) of actual 2D gels (pH range 4 – 7).


a Proteins predicted from sequence does not account for charge b Spots detected in 2D gels from single D. pulex organisms. c Spots detected in 2D gels of a multiple D. magna organisms.

(184.0 ± 2.6 mg or approximately 130 daphnids)

Table 1. Distribution of 2000 most abundant *Daphnia* proteins predicted from genome sequence

Two-Dimensional Gel Electrophoresis Reveals

environments (normal or hypoxic conditions).

**4. Conclusion** 

**5. References** 

331(6017): 555-61.

*Idrobiologia*, 45:11-30.

*Research*, 31(13):3784-3788.

**3.5 Basic proteins constituents of the** *D. magna* **proteome** 

Differential Protein Expression Between Individual *Daphnia* 325

Initially, single Daphnid extracts were analyzed on IPG strips with a pH range of 3-10. Since more than 80% of the proteins visualized by silver staining were in the pH 4-7 range, subsequent analyses were performed using IPGs pH 4-7. However, the theoretical 2D does predict a significant number of proteins (47%) in the basic range (7-10). Due to their relative low abundance, the visualization of very basic proteins (pH 7-10) in Daphnids required many more organisms. For this, 184 ± 3 mg of *D. magna* (approximately 135 organisms) were cultured under either normal or hypoxic conditions and processed in 1.3 mL of ProteoSOLVE IEF Reagent. The samples were concentrated two-fold by ultrafiltration, and IEF was performed on IPGs pH 7-10, followed by silver staining and image analysis. Silver staining detected 355 and 408 spots (gels not shown) in pH range 7-10 from *D. magna* cultured under normal or hypoxic conditions respectively, further illustrating the utility of

Organisms live in ever changing environments. Understanding how individuals respond and adapt to these environments on a molecular level forms the basis for advances in personalized medicine and requires model systems with both well-developed ecological knowledge and molecular resources. The freshwater crustacean *Daphnia* now has these two requirements in place. We have demonstrated the ability of 2-DE to identify protein differences between single *Daphnia magna* with distinct genotypes (Iinb1 and Xinb3), distinct phenotypic differences (presence or absence of ephippia) and when cultured in different

We predict that the detectable proteins on a single animal 2D gel, while biased towards the most abundant proteins, represent a functionally diverse set of proteins. This technique represents an important step to a greater understanding of individual variation of gene expression and is critical to advancing the field of EEFG. However, as the use of silver stain convolutes downstream mass spectrometry, the number of protein spots that can be confidently identified is significantly curtailed. The full potential of single animal gels will be realized with the development of comprehensive 2D maps of the *Daphnia* proteome.

Altschul SF, Gish W, Miller W, Myers EW, & Lipman DJ. 1990. Basic local alignment search

Colbourne J.K., *et. al.,* 2011. The Ecoresponsive Genome of *Daphnia pulex*. *Science*, Feb 4;

Edmondson WT. 1987. *Daphnia* in Experimental Ecology: Notes on Historical Perspectives.

Gasteiger E, Gattiker A, Hoogland C, Ivanyi I, Appel RD & Bairoch A. 2003. ExPASy: The

Geiser HA, Hanneman AJ, & Reinhold V. 2002. HTP proteome-glycome analysis in

In: Peters RH, R. dB, editors. *Daphnia. Pallanza: Memorie dell'Instituto Italiano di* 

proteomics server for in-depth protein knowledge and analysis. *Nucleic Acids* 

tool. *Journal of Molecular Biology*, 215(3):403-410.

Caenorhabditis elegans. *Glycobiology*, 12(10):650.

2-DE for detecting phenotypic differences influenced by specific environments.

Fig. 7. Distribution of highly expressed *D. pulex* genes across 25 KOG classes. Blue and red bars indicate the number of genes in each class from the top 2000 and 1000, respectively, most highly expressed genes (Left-hand axis). Green bars indicate the total genes assigned to each KOG class as a percentage of the total number of genes (18,371) assigned to a KOG class (Right-hand axis). To further characterize the predicted protein gel, we compared it to our observed spot counts. Table 1 summarizes the number of predicted proteins and the number of actual proteins within specific pI ranges. Through the generation of several *D. magna* (both single animal and multiple-animal) 2D gels (not shown), we were able to visualize a total of 1285 protein spots. Seventy percent of these were observed in the 4-7 pI range, while the theoretical gel predicted 51% in this same range. It is important to note that the theoretical 2D does not account for post-translational modifications and was generated from *D. pulex* genes. Both of these factors will contribute to differences between the predicted (*D. pulex*) and observed (*D. magna*) number of proteins.
