**4.4 Separation and characterization of isoforms**

One of the main challenges for human proteomics is to identify and characterize co- and post-translational modifications to be able to study their relevance and place in systems biology. Most human proteins are expressed as different isoforms often depending on posttranslational modifications. Two common modifications in plasma are truncations and glycosylations and here we have used the ability of 2-DE to separate such isoforms based on differences in isoelectric point and molecular mass.

Two-Dimensional Gel Electrophoresis and Mass

in response to endotoxin (Levels et al., 2011).

Spectrometry in Studies of Nanoparticle-Protein Interactions 351

showed that the more acidic isoforms of SAA1 and SAA2 comprised a mixture of truncations with the loss of one, two or four amino acids N-terminally; des-Arg, des-Arg-Ser and des-Arg-Ser-Phe-Phe, respectively, with the loss of arginine being the main explanation to the acidic shifts (figure 13D). On the other hand, the native peptide was only found in the more basic isoforms of SAA1 and SAA2. In total, four forms of SAA1 and four forms of SAA2 was identified. Interestingly, studies of these small molecular mass variants of SAA1 and SAA2 with SELDI-TOF MS indicates population cluster differences in HDL related to the truncations

Fig. 13. Identification of serum amyloid A isoforms by 2-DE and MALDI-TOF MS.

peptide of the native protein (protein spot 2) are indicated.

A: HDL proteins were separated by 2-DE and stained by Sypro Ruby. Arrows indicate the two isoforms of serum amyloid A1 (SAA1) and the two isoforms of serum amyloid A2 (SAA2) identified by peptide mass fingerprinting. B and C: MS spectra after trypsin digestion of SAA1and SAA2, respectively, with specific masses indicated. D: MS spectra after CNBr digestion of SAA1. Masses corresponding peptides from N-terminal truncated variants of the protein (protein spot 1) and the mass corresponding to the N-terminal

2-DE makes it possible to separate proteins based on the degree of glycosylation. Hydrophilic sugars affect the binding of SDS and usually render the proteins an apparent higher molecular mass in the second dimension and the presence of negative sialyl-groups makes the proteins more acidic in the first dimension. We have therefore adapted two simple 2-DE mobility shift assays to demonstrate glycosylation of proteins and applied these to study glycosylated isoforms of plasma proteins in HDL. In the first we use

Serum amyloid A1 and A2 (SAA1 and SAA2, respectively) are two acute phase proteins, which share more than 95 % sequence identity. Both SAA1 and SAA2 can also be expressed as an alpha- and a beta-form, which are discriminated from the others only in one amino acid position (Strachan et al., 1989). SAA1 and SAA2 are associated to HDL (Karlsson et al., 2005) and are heavily induced by endotoxins (Levels et al., 2011), which is highly relevant in particle toxicology to discriminate between different environmental agents (Karlsson et al. 2011). Based on differences in isoelectric points we were able to separate four isoforms of SAA1/2 in HDL (figure 13A). By peptide mass fingerprinting after trypsin digestion we identified two of the isoforms as SAA1 with pI 5.5 and 6 and two as SAA2 with pI 7 and 8 (figures 13A and 13B). The theoretical pI of SAA1 and SAA2 is 5.9 and 8.3, respectively. Thus, the pI of one of the isoforms of SAA1 and of SAA2 corresponded to the theoretical values while the other two had an acidic shift (pI 6→5.5 in SAA1 and pI 8→7 in SAA2). N-terminal truncations of SAA1 and SAA2 that would produce such acidic shifts have previously been described (Ducret et al., 1996) and we therefore focused the MS analyses on the N-terminal peptide. As SAA1 and SAA2 contain arginine at the N-terminus we used CNBr, which cleaves before methionines, as an alternative digestion agent to detect the full length N-terminal peptide. These analyses

Fig. 12. Comparison of α-cyano-4-hydroxycinnamic acid (CHCA) and 2,5-dihydroxybenzoic

Serum amyloid A1 and A2 (SAA1 and SAA2, respectively) are two acute phase proteins, which share more than 95 % sequence identity. Both SAA1 and SAA2 can also be expressed as an alpha- and a beta-form, which are discriminated from the others only in one amino acid position (Strachan et al., 1989). SAA1 and SAA2 are associated to HDL (Karlsson et al., 2005) and are heavily induced by endotoxins (Levels et al., 2011), which is highly relevant in particle toxicology to discriminate between different environmental agents (Karlsson et al. 2011). Based on differences in isoelectric points we were able to separate four isoforms of SAA1/2 in HDL (figure 13A). By peptide mass fingerprinting after trypsin digestion we identified two of the isoforms as SAA1 with pI 5.5 and 6 and two as SAA2 with pI 7 and 8 (figures 13A and 13B). The theoretical pI of SAA1 and SAA2 is 5.9 and 8.3, respectively. Thus, the pI of one of the isoforms of SAA1 and of SAA2 corresponded to the theoretical values while the other two had an acidic shift (pI 6→5.5 in SAA1 and pI 8→7 in SAA2). N-terminal truncations of SAA1 and SAA2 that would produce such acidic shifts have previously been described (Ducret et al., 1996) and we therefore focused the MS analyses on the N-terminal peptide. As SAA1 and SAA2 contain arginine at the N-terminus we used CNBr, which cleaves before methionines, as an alternative digestion agent to detect the full length N-terminal peptide. These analyses

Transthyretin was identified with sequence coverage of 39 % with CHCA (A) versus 72 % with DHB (B). Peptide peaks marked with an asterisk were matched to the theoretical

acid (DHB) as matrices in MALDI-TOF MS of a silver stained protein.

masses with an accuracy <50 ppm.

showed that the more acidic isoforms of SAA1 and SAA2 comprised a mixture of truncations with the loss of one, two or four amino acids N-terminally; des-Arg, des-Arg-Ser and des-Arg-Ser-Phe-Phe, respectively, with the loss of arginine being the main explanation to the acidic shifts (figure 13D). On the other hand, the native peptide was only found in the more basic isoforms of SAA1 and SAA2. In total, four forms of SAA1 and four forms of SAA2 was identified. Interestingly, studies of these small molecular mass variants of SAA1 and SAA2 with SELDI-TOF MS indicates population cluster differences in HDL related to the truncations in response to endotoxin (Levels et al., 2011).

Fig. 13. Identification of serum amyloid A isoforms by 2-DE and MALDI-TOF MS. A: HDL proteins were separated by 2-DE and stained by Sypro Ruby. Arrows indicate the two isoforms of serum amyloid A1 (SAA1) and the two isoforms of serum amyloid A2 (SAA2) identified by peptide mass fingerprinting. B and C: MS spectra after trypsin digestion of SAA1and SAA2, respectively, with specific masses indicated. D: MS spectra after CNBr digestion of SAA1. Masses corresponding peptides from N-terminal truncated variants of the protein (protein spot 1) and the mass corresponding to the N-terminal peptide of the native protein (protein spot 2) are indicated.

2-DE makes it possible to separate proteins based on the degree of glycosylation. Hydrophilic sugars affect the binding of SDS and usually render the proteins an apparent higher molecular mass in the second dimension and the presence of negative sialyl-groups makes the proteins more acidic in the first dimension. We have therefore adapted two simple 2-DE mobility shift assays to demonstrate glycosylation of proteins and applied these to study glycosylated isoforms of plasma proteins in HDL. In the first we use

Two-Dimensional Gel Electrophoresis and Mass

usually form aggregates as shown in this study.

Vol.27, No.2, pp. 156-158

110, ISSN 1097-6744

in toxicological studies.

**6. References** 

5849

Spectrometry in Studies of Nanoparticle-Protein Interactions 353

and generalized conclusions are allowed. Most types of nanoparticles seem to form aggregates, especially so in water suspensions and the "corona" seen might be heavily influenced by the size/diameter of these aggregates rather than by other particle characteristics. Thus, characterization of particles and the aggregates they form prior to exposure of plasma proteins, cells or other biological systems is therefore extremely important. One way of doing that, as we have showed herein, is by DLS. These analyses gives valuable information about the trends in size distribution connected to sample preparation methods and choice of solvent. Sample preparation methods are indeed very important as well as choice of solvent and care should be taken when choosing fitting model and the model-inbuilt parameters. Thus, information obtained from DLS is important for everybody that is doing research on nanoparticles in liquids. The numbers given as product information i.e. the size and size distribution are often relevant for the core-size of the nanocrystals within the material. However the nanoparticles are most often not soluble to that extent. Consequently, nanoparticles obtained in dry state and then dispersed in liquid

Given ample attention to the characterization of the NPs used, future studies of the NP– protein complex behavior in different biological systems are needed. Questions that need to be addressed are which properties of the NPs that govern the protein "corona" formed around the NPs in biological fluids and how these complexes interact with endothelial cells, platelets, cells of the immune system etc. One interesting finding in this study is Amyloid β A4, not previously identified in plasma, which was only associated to ZnO particles. This protein may act as a chelator forming metal-amyloid aggregates and needs further attention

In summary, the improved 2-DE/MS protocols shown herein underline this proteomic approach as a powerful tool in human nano-particle toxicology. Furthermore, thorough characterisation of the particles studied, e.g. with DLS, is crucial to evaluate the results.

Adiseshaiah, P.; Hall, J. & McNeil, S. E. (2010). Nanomaterial Standards for Efficacy and

Beavis R.; Chaudhary T. & Chait B. (1992). α-Cyano-4-hydroxycinnamic acid as a matrix for

Beck-Speier, I.; Dayal, N.; Karg, E.; Maier, K.; Schumann, G.; Schulz, H.; Semmler, M.;

Bell M. & Davis D. (2001). Reassessment of the lethal London Fog of 1952: Novel Indicators

*Environmental Health Perspectives,* Vol.109, pp. 389-394, ISSN 0091-6765 Benderly, M.; Boyko, V. & Goldbourt, U. (2009). Apolipoproteins and Long-Term Prognosis

*Nanobiotechnology,* Vol.2, No.1, pp. 99–112, ISSN 1939-0041

Toxicity Assessment. *Wiley Interdisciplinary Reviews – Nanomedicine and* 

matrixassisted laser desorption mass spectrometry. *Organic Mass Spectrometry*,

Takenaka, S.; Stettmaier, K.; Bors, W.; Ghio, A.; Samet, J. & Heyder, J. (2005). Oxidative stress and lipid mediators induced in alveolar macrophages by ultrafine particles. *Free Radical Biology and Medicine,* Vol.38, No.8, pp. 1080-1092, ISSN 0891-

of Acute and Chronic Consequences of Acute Exposure to Air Pollution.

in Coronary Heart Disease Patients. *American Heart Journal*, Vol.157, No.1, pp. 103–

endoglycosidase PNGase to cleave N-linked oligosaccharides from the protein backbone and the second is based on enzymatic removal of sialic acid with neuraminidase. As shown in figure 14A, SAA4 is usually expressed in HDL as 6 isoforms, three with molecular masses about 18k and three with molecular masses about 11k. After PNGase treatment it was clearly shown that the 18k isoforms are depending on N-linked glycosylation (Fig 14A). Another glycosylated protein in HDL is apo C-III that can be found as three isoforms; one di-sialylated, one mono-sialylated and one minor non-sialylated form (Bruneel et al., 2008). This was demonstrated by treatment with neuraminidase, which induced a mobility shift with the loss of the two sialylated isoforms and a substantial increase of the non-sialyated apo C-III form (figure 14B).

Fig. 14. 2-DE mass and charge mobility shift assays to demonstrate glycosylated protein isoforms.

A: SAA4 analyzed by 2-DE and silver stained. N-linked glycosylated serum amyloid A4 (SAA4) isoforms shown by deglycosylation with PNGase. B: Apo C-III analyzed by 2-DE and Western blots. Sialylated apo C-III isoforms shown by desialylation with neuraminidase.
