Salt mediated unusual switching in the aggregation kinetic profile of human carbonic anhydrase

Abstract

The protein aggregation behaviour is highly dependent on the physico-chemical environment of the protein. Even marginal changes of the solution conditions can severely affect or perturb the aggregation pathway, kinetics and the morphology of the aggregated state. Here, we report an interesting change in the aggregation profile of human carbonic anhydrase II on the addition of salt. Our principal observation is that the salt ions alter the monophasic aggregation kinetics of HCA II to biphasic at 328 K. Solubility studies show that the major portion of soluble protein gets incorporated into aggregates during the first transition. During the second transition, there is just a fusion of pre-formed aggregates resulting in large-sized aggregate clusters. Both the transitions are second order processes, however, the forces responsible for the growth during the two transitions are different as the rate and extent of aggregation are affected differently with increasing salt concentration for two transitions. At the same salt concentration, the cations follow the Hofmeister series in modulating the HCA II aggregation, suggesting a possible role of protein–ion and water–ion interactions along with the changes in general water structure. Temperature dependence studies of the biphasic profile suggest a low and high activation barrier for first and second transition, respectively.

---

1. Introduction

Protein aggregation is one of the major problems during manufacturing, storage, transport and administration of protein-based therapeutics.^1–3 The aggregation of proteins and their deposition into amorphous precipitates or insoluble fibrils can have profound pathological implications.^4–7 In addition, it is a big obstacle during protein expression and purification⁸ and also undesirable during drug delivery.⁹ Thus, it is imperative to understand the mechanisms of protein aggregation and how solution conditions govern this process. In general, the mechanism of protein aggregation can be described in terms of two basic models, namely, nucleation-dependent polymerization and downhill polymerization.^10–14 Several studies have been carried out to understand the mechanism of aggregation; however, understanding of complete picture of mechanism is still elusive.

Environment plays an important part in governing the mechanism, and change in solution conditions has been shown to transform the nucleation dependent polymerization to downhill and vice versa.^15,16 The physico-chemical environment of the protein also exerts a strong influence on the protein aggregation behaviour such as onset, aggregation rate,¹⁷ and the final morphology of the aggregated state (i.e., amorphous precipitates or fibrils).¹⁸ Sometimes unusual behaviour has been observed which challenges our current understanding of the protein aggregation. Several studies are reported where the proteins that are otherwise known to form disordered amorphous aggregates under destabilizing conditions are directed towards amyloid fibrils by modulating the solution conditions.^19–21 Similarly, proteins known to form toxic amyloid fibrils have been directed to form non-toxic amorphous aggregates.²²

One of the important variables, among the series of various factors like pH, temperature, presence of co-solutes^3,23,24 etc., is salt concentration which has been shown to affect self-assembly of protein in diverse ways. Salts occur universally in physiological environment and are known to have profound effect on the solubility of proteins. In several cases such as in aggregation of IgG2 monoclonal antibody and the yeast prion protein Sup35, salts follow Hofmeister series in modulating the aggregation propensity.^25,26 In other cases such as in the case of mouse prion protein and β2-microglobulin, specific anion binding was shown to play an important role in the aggregation process.^27,28 There is also a report where both Hofmeister effect and specific ion binding seems to be crucial.²⁹ Some of the studies have shown that different series are effective in the different concentration regimes of salt.³⁰ Recently, we have seen that the addition of salts like guanidinium hydrochloride (GdmCl) and CaCl₂ changes the aggregation mechanism of bovine serum albumin from downhill to nucleated one.¹⁶

Human carbonic anhydrase II (HCA II) is a member of large zinc metalloenzyme family that catalyses the reversible hydration of carbon dioxide to bicarbonate. It plays an important role in many different physiological processes in animals which include acid–base balance, ion transport, bone resorption, calcification and secretion of gastric, cerebrospinal fluid and pancreatic juices.^31,32 The deficiency of HCA II has been known to be linked with a loss-of-function disease called carbonic anhydrase II deficiency syndrome (CADS).³³ The deficiency of enzyme occurs due to the aberrant protein folding resulting in a population of aggregation-prone intermediate state. Thus, it is worthwhile to study misfolding and aggregation behaviour of CAII which can be helpful in the formulation of viable strategies to deal with this disease.

HCA comprises of a single polypeptide chain, with a molecular mass of 29 kDa and binds one zinc ion in the active site. Structurally, it is composed of single globular domain with a central motif consisting of 10 stranded twisted β sheet.³⁴ Only three short α-helices are present on its surface.³⁵ It has no post-translational modifications and no disulphide bonds,³⁶ and is known to aggregate above its unfolding temperature.³⁷ The presence of large number of β-strands and the molten globule intermediate, make HCA a good model system to investigate the aggregation mechanism of β-sheet rich proteins.³⁶ Since the aggregation of proteins is a generic property, study of the aggregation of such proteins can also provide a wealth of information that can be harvested for insights into other aggregation induced diseases. It can also help in establishing protocol for protein expression and purification and designing strategies for drug delivery.

The aim of the present work is to study the thermally induced aggregation of HCA II in the presence of salt ions. Along with the evaluation of aggregation kinetics, we tried to explore the gross morphological features of HCA II aggregates in a time dependent manner. A well-established right-angle light scattering technique is used to analyse aggregation kinetics behaviour of HCA II and to recognize distinct HCA II aggregates at different time points. To our surprise, addition of salts leads to the transformation of monophasic aggregation pathway of HCA II to more complex biphasic profile. To the best of our knowledge, there is only one case reported wherein protein showed complex, multistep aggregation kinetic behaviour with two sequential transitions.³⁸ Thus, we extended our studies to get better understanding of the underlying phenomenon. We examined the influence of protein concentration, temperature and salt concentration on the biphasic aggregation of HCA II. The efficiency of different cations in accelerating the protein aggregation was evaluated in terms of various possible models to explain their effects, namely the Debye–Huckel screening of charges, specific cation binding (electroselectivity series), and effects on the protein–water–cation interactions (Hofmeister series). The structural and morphological features of isolated time-dependent aggregate species were investigated using additional analytical methods.

2. Experimental section

2.1 Materials

Luria Broth for bacterial cell culture was purchased from Himedia (India). Ampicillin sodium salt was obtained from Biobasic Canada Inc. Isopropyl-β-D-thio-galatose (IPTG) and dithiothreitol (DTT) were purchased from Sisco Research Laboratories Pvt. Ltd. (India). Tris, zinc sulphate and sodium azide were obtained from Merck. Sodium chloride was purchased from Fischer scientific (India). Fluorescent probes 8-anilino-1-naphthalenesulfonic acid (ANS) and nile red (NR) were purchased from Sigma-Aldrich. HCl, H₂SO₄ and NaOH used to maintain pH were also of analytical reagent grade from Sisco Research Laboratories (India). The chemicals were used as such without any further purification. Stock solution of ANS was prepared in Milli Q water, made using Millipore water purification system, and concentration was determined using a molar extinction coefficient of 4900 M⁻¹ cm⁻¹ at 350 nm. Stock solution of nile red was prepared in DMSO and the concentration was determined using a molar extinction coefficient of 19 600 M⁻¹ cm⁻¹ at 552 nm. The pH was determined on a standard Sartorius (PB-11) pH meter at ambient temperature. Unless otherwise mentioned, all solutions were made in 20 mM Tris-buffer (pH 7.5).

2.2 Protein expression and purification

The plasmid containing HCA II cDNA (pACA) was transformed into expression vector, Escherichia coli BL21 (DE3) cells. The transformed E. coli cells harbouring recombinant plasmid were grown in Luria broth at 37 °C containing 100 μg mL⁻¹ ampicillin with shaking to a cell density corresponding to A_{600 nm} ∼ 0.8–1. Once the required O.D. was reached, the temperature of culture flask was brought down to 30 °C and the induction of protein expression was initiated by the addition of 0.25 mM IPTG and 0.5 mM ZnSO₄. After 6 hours of induction, the bacterial culture was harvested by centrifugation at 5000 rpm for 15 min. The pellets were suspended in buffer (50 mM Tris, 50 mM NaCl, 10 mM EDTA, 200 μM ZnSO₄, 1 mM DTT, 1 mM PMSF, pH 8.0) and were lysed by sonication. The purification was carried out using affinity chromatography.³⁹ The cell lysate was loaded onto an agarose resin coupled with p-(aminomethyl)-benzene-sulfonamide, an HCA II inhibitor. The elution of protein was done using 0.4 M sodium azide, which was removed by extensive dialysis against 10 mM Tris–HCl, pH 8.0. Finally, the protein was buffer exchanged into 10 mM Tris–HCl, pH 7.5. Protein purity and integrity was verified by SDS-PAGE (Fig. S1†) and enzymatic activity (PNPA assay), respectively (Fig. S2†). Protein concentration was determined using a molar extinction coefficient of 54 000 M⁻¹ cm⁻¹ at 280 nm on a Cary 100 Bio UV-visible spectrophotometer.

2.3 Aggregation assay

The kinetics of aggregation was monitored at a protein concentration of 0.3 mg mL⁻¹ in the absence or presence of 100 mM NaCl by measuring right angle light scattering with 400 nm as excitation and emission wavelengths at 328 K on a Varian Cary Eclipse Fluorescence Spectrophotometer. For the extrinsic fluorescence studies, the dye was added to the working solution at the beginning and the change in the fluorescence intensity was monitored with time. ANS was excited at 350 nm and the emission was collected at 480 nm, while nile red was excited at 550 nm and the emission was collected at 610 nm. ANS and nile red is generally used to monitor changes in surface hydrophobicity during protein aggregation.^40–42

2.4 Aggregation rate analysis

The observed rate of aggregation was determined by the linear extrapolation of the time-dependent change in the scattering signal within the growth region. The aggregation rate was obtained from the slope of the plot of steepest part of the scattering value change for each aggregation transition versus time.³⁸

2.5 Soluble protein determination and isolation of time-dependent aggregates

HCA II, at a concentration of 0.3 mg mL⁻¹ in Tris buffer, pH 7.5 with 100 mM NaCl, was subjected to an aggregation assay at 328 K. At a specified time, the reaction mixture was removed and immediately placed on ice to quench the aggregation. After 30 min incubation on ice, the aggregation sample was centrifuged (13 000 rpm, 20 min, 10 °C) and the supernatant was carefully transferred into a new 1.5 mL centrifuge tube. The aggregated protein fraction (expressed as a percentage) was determined using the difference in absorbance (A₂₈₀) value between the sample that was not subjected to the aggregation assay (control) and the time point-derived supernatant.

2.6 Dynamic light scattering

The DLS measurements of samples at various incubation times were carried out at 25 °C on a Malvern Zetasizer ZS 90 unit fitted with a 633 nm ‘red’ laser. Before performing measurements, the samples were centrifuged at 13 000 rpm for 20 minutes at 10 °C to remove large aggregates. The sizes reported are average of 30 runs per measurement with correlation time of 10 s for each run.

2.7 Size exclusion chromatography

The samples prepared by quenching the reaction at consecutive time intervals were analysed by size exclusion chromatography using Superdex 200 increase 10/300 GL column in a GE Akta purifier FPLC system using a loop of 1 mL and an injection volume of 900 μL. Before injection, the samples were filtered using a 0.22 μm syringe filter to remove any higher or insoluble aggregates, to prevent clogging of the column. Samples were eluted isocratically and the mobile phase used was 20 mM Tris, 100 mM NaCl at pH 7.5.

2.8 Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR)

The FTIR spectra of native protein, and soluble protein isolated at time points 3 and 5 were obtained using Cary 600 series FTIR spectrometer (Agilent Technologies). Samples were applied to a germanium crystal in the ATR sampling accessory, and then spectra were recorded at a resolution of 4 cm⁻¹. For each sample, 128 interferograms scans were averaged. Buffer baseline was recorded under identical conditions and was used as a blank before taking the respective spectra. The shift in position of band in FTIR spectra is generally used to observe changes in secondary structure of protein during aggregation.⁴³

2.9 Scanning electron microscopy

Electron microscopy has emerged as a powerful tool to visualize the type and morphology of species formed during aggregation process.⁴⁴ Samples prepared for five different points were centrifuged at 13 000 rpm, 10 °C for 20 min. The supernatant was discarded and the pellet containing aggregates was thoroughly washed with cold Milli Q water to get rid of any soluble protein and salt and resuspended with 1 × volume of Milli Q water. A sample volume of 10 μL for each time point preparation was drop-casted on a glass coverslip. The droplet was spread across the surface of the coverslip, using the pipette tip, and was allowed to air-dry. SEM images were recorded using ZEISS EVO series scanning electron microscope EVO 50 operating at an accelerating voltage of 0.2–30 kV. Control experiments with the native protein alone did not show any aggregates.

2.10 Transmission electron microscopy

TEM images of samples were acquired using a Tecnai G2 200 kV transmission electron microscope (FEI Co.). Preparation of aggregate sample was same as that of SEM. 10 μL of diluted sample for time point 3 and 5 was deposited on carbon coated copper grids and allowed to adsorb for 5 min followed by staining with 1% (w/v) phosphotungstic acid solution. The grid was air-dried for about 30 min and examined at a voltage of 80 kV.

3. Results and discussion

3.1 Monophasic kinetic profile of human carbonic anhydrase II (HCA II) aggregation at 328 K

Protein aggregation was monitored by right-angle light scattering at high temperature (328 K) using fluorescence spectroscopy. The aggregation kinetic trace of human carbonic anhydrase, a β-sheet rich protein, displayed a typical single-transition profile consisting of a lag phase, followed by phases of linear growth and plateau (Fig. 1A), suggesting a nucleation dependent polymerization. 1-Anilinonaphthalene-8-sulphonate (ANS) was used to monitor the accessibility of protein hydrophobic patches during the course of aggregation. At 328 K, in the absence of 100 mM NaCl, HCA II showed immediate increase in ANS intensity without any lag phase (Fig. 1B). A similar kinetic profile lacking the lag phase was seen when the aggregation was monitored by nile red, another important dye that binds to hydrophobic surfaces of proteins (Fig. S3†). A time scan of the tryptophan fluorescence displayed a concomitant decrease in the fluorescence intensity, indicating the exposure of tryptophan residues to a more solvent accessible state with time (Fig. S4†). All of these observations suggest that the conformational changes proceed without a lag phase before the onset of the formation of any detectable protein aggregates. Also, the aggregation of HCA could not be captured by ThT fluorescence implying that the protein aggregates formed under above-mentioned condition are amorphous in nature, not amyloid (data not shown). This is in contrast to aggregation of HCA in the presence of trifluoroethanol (TFE), where amyloid formation has been seen.²¹

Fig. 1 Aggregation kinetics of HCA II (0.3 mg mL) which was initiated by heating at 328 K in 20 mM Tris–H₂SO₄ buffer (pH 7.5) without NaCl. Aggregation was monitored by time dependent changes in the (A) right angle light scattering intensity at 400 nm (B) ANS fluorescence intensity at 480 nm.

3.2 Addition of salt changes monophasic aggregation kinetic profile to biphasic for human carbonic anhydrase II at 328 K

Interestingly, on addition of salt ions, the biphasic aggregation kinetic behaviour with two distinct transitions and a remarkable decrease in the overall aggregation was observed. Fig. 2A shows an aggregation profile of 0.3 mg mL⁻¹ HCA II at 328 K in the presence of 100 mM NaCl. At this salt concentration, protein aggregation showed an initial lag phase, accompanied by a sharp growth phase corresponding to the formation of protein aggregates. The plateau of first transition corresponds to small lag phase of forthcoming second transition. No lag phase and biphasic behaviour was observed when aggregation process was monitored using ANS, a hydrophobic probe, in the presence of salt (Fig. 2B).

Fig. 2 Aggregation kinetics of HCA II (0.3 mg mL⁻¹) upon heating at 328 K in 20 mM Tris–H₂SO₄ buffer (pH 7.5). Aggregation was monitored by time dependent changes in the (A) right angle light scattering intensity at 400 nm in the absence () and presence () of 100 mM NaCl. (B) ANS fluorescence intensity at 480 nm.

The amount of soluble protein remaining in the solution at different time points was determined by taking absorbance at 280 nm. This fraction was then subtracted from total initial protein to estimate the amount of aggregated protein. The results show that the first transition is responsible for the aggregation of most of the protein (Fig. 3) and there is just an increase in the size of pre-formed aggregates during the second transition without the recruitment of soluble protein from the solution.

Fig. 3 Aggregation kinetic profile of HCA II at 328 K in the presence of 100 mM NaCl () and corresponding estimates of aggregated protein ().

3.3 Effect of increasing salt concentration on the aggregation of HCA II at pH 7.5 and 328 K

The aggregation kinetic profile of HCA II was found to be highly sensitive to NaCl concentration. Fig. 4 shows the dependence of HCA II aggregation on the salt concentration (0–300 mM NaCl) at 328 K and pH 7.5. In our study, it was seen that the addition of salt decreases the aggregation propensity of HCA II in a concentration dependent manner. This observation suggests the importance of electrostatics in the aggregation of the protein. The presence of salt minimizes the intermolecular attractive forces between protein molecules which may be due to the screening effect of salt or the binding of ions to the protein's multiple ionic charges. This keeps the protein more in the soluble form, which in turn leads to overall reduced aggregation. Also, a clear transition from monophasic to biphasic aggregation profile was observed beyond 10 mM of NaCl concentration; the transition becomes more prominent with further increase in salt concentration. In general, it is believed that screening of electrostatic repulsion is prerequisite for aggregation of all protein. In this case, suppression of aggregation at high ionic strength arises from reduction in favorable attractive interaction due to screening effect.

Fig. 4 Effect of salt concentration on the aggregation of HCA II at 328 K. Aggregation was monitored by time dependent change in the scattering intensity at 400 nm. The curves denote different salt (NaCl) concentrations (a) 0 mM () (b) 10 mM () (c) 25 mM () (d) 50 mM () (e) 100 mM () (f) 300 mM ().

Although overall protein aggregation was reduced with increasing salt concentration, but the aggregation rate of two transitions was found to be affected differently. The first transition shows a remarkable change in terms of both the rate and extent of aggregation with increasing salt concentration, whereas rate of second transition was found to be affected insignificantly with salt concentration (Fig. S5†). This observation suggests that the forces controlling the two growth phases are different. The aggregation rate drops drastically beyond 50 mM NaCl during first transition.

3.4 Effect of protein concentration and temperature on the aggregation of HCA II in the presence of 100 mM NaCl at pH 7.5

3.4.1 Effect of protein concentration. The kinetics of the thermal aggregation of HCA II was studied at different protein concentration in the presence of 100 mM NaCl at 328 K. The biphasic nature of aggregation profile was observed under certain range of protein concentration. In the range of 0.2–0.5 mg mL⁻¹ protein, aggregation shows a double transition phenomenon (Fig. 5A), below and above which it changes to monophasic (Fig. S6†). The lag phase of first (type I) transition phase became shorter with increasing protein concentration, suggesting a faster nucleation phenomenon. This was followed by an overall faster first transition, a shorter first-transition termination phase (equal to the type II lag phase), and finally, a faster second transition.

Fig. 5 Effect of protein concentration on the biphasic aggregation of HCA II in the presence of 100 mM NaCl at 328 K. (A) Aggregation was monitored by time dependent change in the scattering intensity at 400 nm. The curves denote HCA II (a) 0.2 mg mL⁻¹ () (b) 0.3 mg mL⁻¹ () (c) 0.4 mg mL⁻¹ () (d) 0.5 mg mL⁻¹ (). (B) Effect of protein concentration on HCA II aggregation rates. First () and second () kinetic transition data are plotted and linearly fitted, showing slope values of 2.40 and 2.26, respectively.

Fig. 5B shows the aggregation-rate analysis of HCA II assayed at different protein concentrations. The aggregation kinetic rates of both the transitions increased linearly with increasing concentration from 0.2–0.5 mg mL⁻¹. The slope of the linear fitted log–log plot of rate of aggregation versus concentration was used to determine the order of the reaction. Under our experimental conditions, both the transitions followed a second-order reaction process.

3.4.2 Effect of temperature. Fig. 6A depicts the dependence of HCA aggregation on temperature in the presence of 100 mM NaCl. Our temperature studies were restricted to the region 324–330 K as below 324 K no biphasic aggregation pattern was observed even on prolonged incubation (Fig. S7A†) and above 330 K aggregation is too fast such that second lag phase could not be seen properly. In fact, at 343 K and above, the two transitions merge making second lag phase undetectable (Fig. S7B†). The aggregation rates and length of lag phase for both the transitions were found to increase and decrease, respectively, with increasing temperature. It is clear from the Fig. 6B that the aggregation rate increases exponentially with temperature for both the transitions.

Fig. 6 Effect of temperature on the biphasic aggregation of HCA II in the presence of 100 mM NaCl. (A) Aggregation was monitored by time dependent change in the scattering intensity at 400 nm. The curves denote HCA II at (a) 324 K () (b) 326 K () (c) 328 K () (d) 330 K (). (B) Observed aggregation rates (unit per min) for first () and second () transitions of aggregation are plotted against temperature (K). Protein concentration is 0.3 mg mL⁻¹. (C) Arrhenius plot of aggregation rates for two transitions versus inverse temperature. The data for two transitions were linearly fitted and the values of activation energies for the elongation process are derived.

The plot of ln(k) versus 1/T (Fig. 6C) gives a straight line, suggesting that the data follow Arrhenius' equation, k = A exp(−E_a/RT), where A, T, E_a and R are the pre-exponential factor, the absolute temperature (K), the activation energy of the reaction and the ideal gas constant, respectively. From the Arrhenius plot, the activation energies associated with the growth of aggregates for first and second transition were calculated to be 96.97 kcal mol⁻¹ and 208.39 kcal mol⁻¹, respectively. The pre-exponential factor which depends upon the collision frequency and the orientation of molecules when they collide has a natural logarithmic value of 151.53 and 321.32 for first and second transition, respectively (Table 1).

Table 1 Parameters for the formation of transition state during growth of HCA II aggregates for two transitions

	Ea (kcal mol^−1)	ln A
Transition I	96.97 ± 5.2	151.53 ± 7.5
Transition II	208.39 ± 12.8	321.32 ± 16.1

---

3.5 Effect of different cations on the aggregation propensity of HCA II at pH 7.5

HCA II (theoretical pI 6.8) carries negative charge under the pH conditions of the present study; therefore, interactions with positively charged ions are likely. Hence, we monitored and compared the effect of various salts such as NaCl, KCl, NH₄Cl and MgCl₂ (composed of different cations and the same counter anion, Cl⁻) on the aggregation kinetics of HCA II at pH 7.5 (Fig. 7). Different chloride salts were shown to have different potency by which they promote aggregation. Three major possibilities known to explain the effect of salt ions on the protein aggregation: Debye–Huckel screening, electroselectivity series or changes to the water structure (Hofmeister series) has been explored to explain our observations.

Fig. 7 Effect of different cations (100 mM) on the aggregation kinetics of HCA II at pH 7.5 and 328 K. The curves denote different salt ions (a) NH₄Cl () (b) NaCl () (c) KCl () (d) MgCl₂ ().

If the Debye–Huckel screening was the major contributor of protein aggregation, then the effect of various cations on the observed growth of aggregation of HCA would have been determined by the ionic strength of the solution. Since 100 mM salt concentration was used, the calculated ionic strength for the solutions of mono- and divalent cations was 100 and ∼300 mM, respectively. It was observed that the presence of equivalent ionic strength of NaCl and NH₄Cl produces significant difference in their effects on the aggregation of HCA II. Also, MgCl₂, despite providing maximum ionic strength shows least aggregation (Fig. 7). Thus, we conclude that the primary effect of cations in modulating HCA II aggregation is not due to Debye–Huckel screening of protein charges.

Alternatively, if salt act by direct cation–peptide interaction then the effect was expected to follow the electroselectivity series. According to the electroselectivity series, the order of cations for binding to the anion-exchange resin is (from stronger to weaker binders to the anion-exchange resin):

Mg²⁺ > K⁺ > Na⁺ > Li⁺

However, the order of cations in their ability to promote the protein aggregation in our study was as follows:

NH₄⁺ >Na⁺ ≥ K⁺ > Mg²⁺

that is almost reverse of the electroselectivity series of ion binding. This observation ruled out the role of specific anion binding in determining the effect of cations on HCA II aggregation.

Instead, the currently established series follows the Hofmeister series of cations suggesting the important role of ion–water and peptide–water interaction during aggregation process. Protein aggregation data correlate well with the thermodynamic measurements. NH₄⁺ ion was found to stabilize the native protein and showed the maximum salting-out tendency, whereas Mg²⁺ ion acted as chaotrope (denaturant) with maximum salting-in propensity (Fig. S8†).

Salting-out phenomena is not observed at high salt concentration. Further, both chaotropic and kosmotropic salts inhibit the rate as well as the extent of aggregation. This is opposite to the cases where salts do follow Hofmeister series but enhance the rate of aggregation.

3.6 Structural changes associated with HCA II upon aggregation in the presence of 100 mM NaCl at 328 K

Bio analytical techniques like dynamic light scattering (DLS) and size exclusion chromatography (SEC) are widely used to get insights into the nature and sizes of soluble aggregates.⁴⁵ Hence, we analysed the soluble fraction of protein isolated at five consecutive time points (Fig. 8) during the course of the reaction using dynamic light scattering (DLS) and size exclusion chromatography (SEC) to look at the changes in the molecular dimension/size of HCA II aggregates.

Fig. 8 The biphasic kinetic plot indicating the five different time points from which aggregates were collected for further analyses.

Fig. 9 shows the DLS profiles of different soluble species formed during the course of aggregation. DLS studies reveal that the presence of both small (1–10 nm) and large sized (100–800 nm) soluble species increases with time. At the saturation phase of second transition (TP 5), almost all the protein gets aggregated, so no particles of small molecular size (1–10 nm) were detected. Also, broadening of peaks was observed at time points 3 and 4 indicating the formation of species of heterogeneous size at the respective time intervals.

Fig. 9 Time evolution of the distribution of particle size of small (left panel) and large sized aggregates (right panel) at 328 K in the presence of 100 mM NaCl.

Size exclusion chromatography was employed to identify the various soluble forms (dimer, tetramer, or higher species) of aggregates. Native HCA II in the absence of any additive was eluted at 16.86 mL in size exclusion chromatography. Fig. 10a shows the size exclusion profiles of soluble species present in the reaction mixture at different time points over the aggregation process. Since the protein concentration for samples corresponding to time points 1 and 2 were going out of the detection limit of the instrument, we diluted the samples and ran them separately as shown in Fig. 10b. No shift in the elution volume at TP 1 and TP 2 were seen which indicates that protein is present in monomeric form. Interestingly, a peak shift was observed at lower elution volumes for TP 3 and TP 4; protein at TP 3 being eluted at 15.862 mL and TP 4 at 15.782 mL respectively. This indicates an increase in both the number and the size of the aggregates. With further increase in time, the elution peak was shifted back towards the peak of native monomeric protein. This indicates that at TP 5 most of the soluble species gets converted into higher insoluble aggregates and removed during centrifugation, and only monomeric soluble protein was left in the solution giving a small peak at 16.34 mL.

Fig. 10 SEC elution profile of different soluble species of HCA II at time points 2–5. Inset shows the profile for native HCA II (A). SEC profile of diluted sample of time points 1 and 2 (B).

The structural changes associated with soluble fraction of protein isolated at time points 3 and 5 were analysed using FTIR. The FTIR spectrum of native HCA II shows a peak around 1635 cm⁻¹, which is suggestive of the presence of intramolecular β-sheet structure (Fig. 11). The soluble species at TP 3 and 5 also shows a peak around 1620–1640 cm⁻¹ region, suggesting that they are both rich in β-structures. But the position of the peak for the samples at TP 5 is shifted from TP 3, which indicates that these species differ in the internal structures of their β-sheets compared to the native protein.

Fig. 11 ATR-FTIR spectra of native protein and soluble species isolated at time points 3 and 5.

3.7 Morphological properties of HCA II aggregates

To get deeper insight into protein aggregate populations, we examined the morphological properties of insoluble aggregates collected at five different time points using SEM, and results are shown in Fig. 12. All aggregates were washed thoroughly and made free of soluble protein and salt before examination by SEM. The aggregates at the time point 1 and 2 were highly fused small spherical particles, almost indistinguishable from each other making the particle size analysis difficult. As the aggregation proceeds, the spheres continue to grow, possibly because of the independent growth or the fusion of smaller spheres. The aggregate size reached 0.4–0.5 μm in diameter by the end of first transition (time point 3). The second transition aggregates showed further enhancement in size, and in some cases reached up to 1 μm in size. At time points 3, 4 and 5, the spheres became well-developed, but remain interconnected and existed as chains of fused particles within the aggregate clump. TEM micrographs also showed the development of spherical particles as highly fused entities with time (Fig. S9†).

Fig. 12 Scanning electron micrographs of HCA II aggregates collected at time points 1–5 (A–E). Magnification, ×20 000. (A and B) Time points 1 and 2 represent type I transition. (D and E) Time points 4 and 5 represent type II transition, and demonstrate aggregate growth up to 0.95 μm diameter. (C) Time point 3 represents midpoint between transitions. Scale bars are 200 nm for (A), (B) and (E) and 1 μm for (C) and (D).

4. Conclusions

This study describes the effect of salt on the heat induced aggregation kinetics of HCA II. We have shown that presence of salt ions could reshape the monophasic aggregation pathway of HCA II to the more complex, multistep biphasic one, with two distinct transitions at 328 K. Both the kinetic transitions are sensitive to even slight changes in temperature and salt concentration, supporting the role of hydrophobic and electrostatic interactions. We are also able to calculate the activation energy of the two transitions. The activation energy of second transition is quite higher in comparison to activation energy for first transition. Thus, shifting from first transition to second transition has an important advantage.

The aggregation study with different cationic salts follows the Hofmeister series, indicating a relevance of the water structure. MgCl₂ tend to facilitate protein denaturation and increase solubility whereas NH₄Cl stabilize the native folded structure of a protein and facilitate aggregation.

Fig. 13 shows the most probable aggregation pathway for HCA in the presence of salt. Most of the soluble protein gets assembled into aggregates during the first transition as seen from solubility studies. During second kinetic transition, there is a fusion of existing spheres resulting in large spherical particles and less packed aggregate clusters without the incorporation of soluble protein from the solution. SEM data also indicates the assemblage of protein into small, spherical particles coalesced with each other. To summarize, our data reveal the modulation of aggregation pathway of HCA II by salt ions, going from type I to type II transition.

Fig. 13 Schematic representation of proposed HCA II aggregation pathway at elevated temperatures (324–330 K) in the presence of salt. Monomer and putative oligomers are shown as dark blue and light blue circles, respectively.

The general mechanism of elongation for amyloid/aggregate involves the sequential addition of monomer to the aggregate.⁴⁶ Here we show that HCA aggregates in biphasic manner in the presence of salt. Initially, aggregation appears to happen due to the binding of monomer to oligomer (small type I aggregate) followed by oligomer binding to another oligomer/aggregate forming fully developed type II aggregates. This has important consequences since binding of an inhibitor to oligomer may not affect the elongation in first case (monomer binding to aggregate), however it can affect the elongation in second case (oligomer binding to oligomer/aggregate). On the other hand, binding of an inhibitor to monomer may not affect the elongation in second case. This can be the probable reason for the differential effect of salt on the elongation process of two observed transitions. The idea of biphasic aggregation thus provides one an opportunity to develop drugs that can target the different species formed during the aggregation pathway individually.

The physiological interior of a cell is composed of many ions having different valencies. The salts have shown to affect the types of aggregates. Since the different types of aggregates are known to have different toxicity, this study suggests that the ions themselves are capable of modulating the type and morphology of the aggregates and may modulate the toxicity of aggregates.

The observed influence of the ions can be quite advantageous from a synthetic point of view. Peptide or protein-based molecular assemblies of late have found widespread usage with regards to hydrogels.⁴⁷ Hydrogels are increasingly used in tissue engineering, drug delivery scaffolds, biomedical application and biotechnology. Thus one can potentially use these ions to control the self-assembly of the peptides depending on the nature of aggregate in solution, the latter being dictated by the presence of ions present.

Acknowledgements

This work is generously supported by the Council for Scientific and Industrial Research (CSIR), Government of India through grants to S. D. and research fellowship to P. G. We thank Dr Carol Fierke (University of Michigan, Ann Arbor, Michigan) for providing us the plasmid containing the full length coding region of Carbonic Anhydrase II gene. We also acknowledge Central SEM and TEM facility (IIT Delhi) for obtaining micrographs. We thank Dr Pramit K Chowdhury (Department of Chemistry, Indian Institute of Technology, Delhi) for invaluable suggestions during revision of the manuscript.

References

1. J. F. Carpenter, B. S. Kendrick, B. S. Chang, M. C. Manning and T. W. Randolph, Methods Enzymol., 1999, 309, 236–255.
2. W. Wang, Int. J. Pharm., 1999, 185, 129–188.
3. W. Wang, S. Nema and D. Teagarden, Int. J. Pharm., 2010, 390, 89–99.
4. C. M. Dobson, Philos. Trans. R. Soc. London, Ser. B, 2001, 356, 133–145.
5. P. T. Lansbury Jr, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 3342–3344.
6. E. H. Koo, P. T. Lansbury Jr and J. W. Kelly, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 9989–9990.
7. J. Hardy and D. J. Selkoe, Science, 2002, 297, 353–356.
8. S. E. Bondos and A. Bicknell, Anal. Biochem., 2003, 316, 223–231.
9. W. Wang, S. K. Singh, N. Li, M. R. Toler, K. R. King and S. Nema, Int. J. Pharm., 2012, 431, 1–11.
10. W. F. Xue, S. W. Homans and S. E. Radford, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 8926–8931.
11. M. F. Bishop and F. A. Ferrone, Biophys. J., 1984, 46, 631–644.
12. A. Scire, M. Baldassarre, R. Galeazzi and F. Tanfani, Biochimie, 2013, 95, 158–166.
13. S. Kumar and J. B. Udgaonkar, Curr. Sci., 2010, 98, 639–656.
14. A. R. Hurshman, J. T. White, E. T. Powers and J. W. Kelly, Biochemistry, 2004, 43, 7365–7381.
15. E. T. Powers and D. L. Powers, Biophys. J., 2006, 91, 122–132.
16. S. Saha and S. Deep, J. Phys. Chem. B, 2014, 118, 9155–9166.
17. P. R. Majhi, R. R. Ganta, R. P. Vanam, E. Seyrek, K. Giger and P. L. Dubin, Langmuir, 2006, 22, 9150–9159.
18. W. S. Gosal, I. J. Morten, E. W. Hewitt, D. A. Smith, N. H. Thomson and S. E. Radford, J. Mol. Biol., 2005, 351, 850–864.
19. V. Vetri, C. Canale, A. Relini, F. Librizzi, V. Militello, A. Gliozzi and M. Leone, Biophys. Chem., 2007, 125, 184–190.
20. Y. Yoshimura, Y. Lin, H. Yagi, Y. H. Lee, H. Kitayama, K. Sakurai, M. So, H. Ogi, H. Naiki and Y. Goto, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 14446–14451.
21. P. Gupta and S. Deep, Biochem. Biophys. Res. Commun., 2014, 449, 126–131.
22. N. K. Bhatia, A. Srivastava, N. Katyal, N. Jain, M. A. Khan, B. Kundu and S. Deep, Biochim. Biophys. Acta, 2015, 1854, 426–436.
23. E. Y. Chi, S. Krishnan, T. W. Randolph and J. F. Carpenter, Pharm. Res., 2003, 20, 1325–1336.
24. A. Sharma, J. M. Pasha and S. Deep, J. Colloid Interface Sci., 2010, 350, 240–248.
25. V. Yeh, J. M. Broering, A. Romanyuk, B. Chen, Y. O. Chernoff and A. S. Bommarius, Protein Sci., 2010, 19, 47–56.
26. P. Arosio, B. Jaquet, H. Wu and M. Morbidelli, Biophys. Chem., 2012, 168–169, 19–27.
27. B. Raman, E. Chatani, M. Kihara, T. Ban, M. Sakai, K. Hasegawa, H. Naiki, M. Rao Ch and Y. Goto, Biochemistry, 2005, 44, 1288–1299.
28. S. Jain and J. B. Udgaonkar, Biochemistry, 2010, 49, 7615–7624.
29. K. Klement, K. Wieligmann, J. Meinhardt, P. Hortschansky, W. Richter and M. Fandrich, J. Mol. Biol., 2007, 373, 1321–1333.
30. M. Bostrom, D. F. Parsons, A. Salis, B. W. Ninham and M. Monduzzi, Langmuir, 2011, 27, 9504–9511.
31. W. R. Chegwidden, S. J. Dodgson and I. M. Spencer, EXS, 2000, 343–363.
32. M. I. Hassan, B. Shajee, A. Waheed, F. Ahmad and W. S. Sly, Bioorg. Med. Chem., 2013, 21, 1570–1582.
33. K. Almstedt, M. Lundqvist, J. Carlsson, M. Karlsson, B. Persson, B. H. Jonsson, U. Carlsson and P. Hammarstrom, J. Mol. Biol., 2004, 342, 619–633.
34. U. Carlsson and B. H. Jonsson, EXS, 2000, 241–259.
35. S. Lindskog, Pharmacol. Ther., 1997, 74, 1–20.
36. V. M. Krishnamurthy, G. K. Kaufman, A. R. Urbach, I. Gitlin, K. L. Gudiksen, D. B. Weibel and G. M. Whitesides, Chem. Rev., 2008, 108, 946–1051.
37. R. Lavecchia and M. Zugaro, FEBS Lett., 1991, 292, 162–164.
38. S. Krishnan and A. A. Raibekas, Biophys. J., 2009, 96, 199–208.
39. J. F. Domsic, B. S. Avvaru, C. U. Kim, S. M. Gruner, M. Agbandje-McKenna, D. N. Silverman and R. McKenna, J. Biol. Chem., 2008, 283, 30766–30771.
40. N. K. Bhatia and S. Deep, Journal of Proteins and Proteomics, 2013, 4, 149–164.
41. A. Hawe, M. Sutter and W. Jiskoot, Pharm. Res., 2008, 25, 1487–1499.
42. M. Lindgren, K. Sorgjerd and P. Hammarstrom, Biophys. J., 2005, 88, 4200–4212.
43. J. Kong and S. Yu, Acta Biochim. Biophys. Sin., 2007, 39, 549–559.
44. A. E. Langkilde and B. Vestergaard, FEBS Lett., 2009, 583, 2600–2609.
45. J. den Engelsman, P. Garidel, R. Smulders, H. Koll, B. Smith, S. Bassarab, A. Seidl, O. Hainzl and W. Jiskoot, Pharm. Res., 2011, 28, 920–933.
46. L. Nielsen, S. Frokjaer, J. Brange, V. N. Uversky and A. L. Fink, Biochemistry, 2001, 40, 8397–8409.
47. A. M. Jonker, D. W. P. M. Lowik and J. C. M. van Hest, Chem. Mater., 2012, 24, 759–773.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra17794d

---