Quantitative Studies on the Interaction between Saposin-like Proteins and Synthetic Lipid Membranes

Members of the saposin-fold protein family and related proteins sharing a similar fold (saposin-like proteins; SAPLIP) are peripheral-membrane binding proteins that perform essential cellular functions. Saposins and SAPLIPs are abundant in both plant and animal kingdoms, and peripherally bind to lipid membranes to play important roles in lipid transfer and hydrolysis, defense mechanisms, surfactant stabilization, and cell proliferation. However, quantitative studies on the interaction between proteins and membranes are challenging due to the different nature of the two components in relation to size, structure, chemical composition, and polarity. Using liposomes and the saposin-fold member saposin C (sapC) as model systems, we describe here a method to apply solution NMR and dynamic light scattering to study the interaction between SAPLIPs and synthetic membranes at the quantitative level. Specifically, we prove with NMR that sapC binds reversibly to the synthetic membrane in a pH-controlled manner and show the dynamic nature of its fusogenic properties with dynamic light scattering. The method can be used to infer the optimal pH for membrane binding and to determine an apparent dissociation constant (KDapp) for protein-liposome interaction. We propose that these experiments can be applied to other proteins sharing the saposin fold.


Introduction
Saposins, also known as SAPs (sphingolipid activator proteins), are lysosomal proteins that activate enzymes involved in lipid degradation [1]. Saposins exist in soluble and membrane-bound forms; the latter resulting from peripheral membrane binding [2]. Saposins give name to the saposin fold characterized by four to five alpha helices, six conserved cystines, and a V-shaped open membrane-bound form [3]. The saposin fold is shared by other proteins known as saposin-like proteins (SAPLIPs) displaying a variety of functions that require peripheral membrane binding [4]. SAPLIPs have the capability to dimerize in the presence of micelles and liposomes [3] and some have been shown to induce liposome fusion [5,6]. SAPLIPs exhibit diverse amino acid sequences and are found in different eukaryotic systems performing functions that include sphingolipid catabolism, surfactant stabilization, and antimicrobial activity [4,5].
SAPLIPs have relevant scientific and medical applications beyond their native biological functions. For example, human saposins are used to form lipid nanoparticles [7] that are currently utilized to stabilize transmembrane proteins for high-resolution threedimensional structural studies [8,9]. NK-lysin is another member of the saposin fold with cytotoxic activity; peptides derived from NK-lysin's amino acid sequence show antiparasitic properties against P. dicentrarchi via membrane disruption of the parasitic protozoa [10]. Surfactant protein B (SP-B), which facilitates respiration in the lungs by reducing the surface tension at the air-liquid interface, also shares the saposin fold [11,12]. However, SP-B Figure 1. SapC undergoes a pH-dependent conformational change upon peripheral binding to liposomes. (a) Ribbon diagrams of soluble (closed form, left) and membrane-bound (V-shape form, right) sapC conformations determined by high-resolution NMR [2,3]. Arrows indicate reversibility with pH in binding and conformational change. (b) SapC in open conformation binds to the liposome surface.

Experimental Design
The main experimental stages of this method include protein expression and purification, production of liposomes, preparation of NMR samples, NMR fast-acquisition experiments for the titration of protein solutions at micromolar concentration with liposomes, and preparation of samples for fusogenic experiments using dynamic light scattering.
A schematic representation of the connection between NMR signal intensity and protein-liposome binding is represented in Figure 2. Protein-lipid interactions are traditionally difficult to study by biophysical techniques. For high-resolution NMR, limitations result from particle size due to signal broadening. Large particles, such as liposomes, tumble slowly in solution thus reducing the NMR signalto-noise ratio to the point where the NMR signals are effectively 'invisible' [2,6]. Watersoluble, peripheral membrane binding proteins can be studied by NMR in the unbound conformation. Once the protein binds to liposomes, the resulting proteoliposomes tumble at a rate characteristic of the liposome size, thus turning the protein 'invisible' by NMR.
It is reasonable to question whether proteins that bind to liposomes and are capable of lateral movement within the lipid bilayer will show rotational correlation times close to those of the proteins free in solution. To facilitate understanding this concept, the work by Wand et al. on protein NMR in reverse micelles could serve as an ideal example [17].
Extensive evidence from this work shows that the rotational correlation time of proteins tumbling freely in aqueous solutions enclosed by reverse micelles significantly decreases due to the very fast tumbling rate of the reverse micelle [17]. Thus, proteins will show an overall tumbling rate close to that of the particle to which they bind or within which they are enclosed, even if they still undergo translational or rotational movement. Therefore, when considering differences in size such as those described here for SAPLIPs and liposomes, Methods Protoc. 2022, 5, 19 3 of 15 proteins that bind to liposomes will show an overall tumbling rate close to that of the liposome, even if the proteins are freely moving in the lipid bilayer.
It follows that in the presence of an equilibrium between unbound and liposomebound protein, only the former will contribute to the NMR signal intensity. If the equilibrium shifts toward the bound form, the NMR signal intensity will decrease as the fraction of free protein decreases. This effect in solution NMR spectroscopy can be leveraged to perform quantitative studies on the binding of proteins to liposomes or membranes. Here, we describe the application of this method using sapC-liposome binding as a model that can analogously be applied to other SAPLIPs. Furthermore, we show that the method is applicable to identify changes in liposome binding affinity of sapC mutants designed to improve membrane interactions.
Finally, we describe a method using dynamic light scattering in real time to test the fusogenic capabilities of sapC and several mutants, and to determine the time required for liposome fusion.

Experimental Design
The main experimental stages of this method include protein expression and purification, production of liposomes, preparation of NMR samples, NMR fast-acquisition experiments for the titration of protein solutions at micromolar concentration with liposomes, and preparation of samples for fusogenic experiments using dynamic light scattering.
A schematic representation of the connection between NMR signal intensity and protein-liposome binding is represented in Figure 2. SapC undergoes a pH-dependent conformational change upon peripheral binding to liposomes. (a) Ribbon diagrams of soluble (closed form, left) and membrane-bound (V-shape form, right) sapC conformations determined by high-resolution NMR [2,3]. Arrows indicate reversibility with pH in binding and conformational change. (b) SapC in open conformation binds to the liposome surface.

Experimental Design
The main experimental stages of this method include protein expression and purification, production of liposomes, preparation of NMR samples, NMR fast-acquisition experiments for the titration of protein solutions at micromolar concentration with liposomes, and preparation of samples for fusogenic experiments using dynamic light scattering.
A schematic representation of the connection between NMR signal intensity and protein-liposome binding is represented in Figure 2.   Figure 2. Schematic representation of the effect of liposome binding on NMR signal intensity. NMR signal intensity from soluble sapC in the absence of liposomes (dark blue) and residual intensity in the presence of liposomes resulting from~5% sapC free in solution (light blue).

•
Plasmids for the following constructs were obtained from Gene Universal in the pET-30b vector with a C-terminal six-Histidine tag following a thrombin cleavage site. The following constructs were used in these studies: sapC, sapC-PUMA and sapC-PUMA-DM [6]. The latter construct is a double-mutant (DM) of sapC-PUMA for enhanced protein-liposome binding [ • Divide the surface area of the membrane by the surface area of the head group (0.64 nm 2 ) to determine the total number of lipid molecules in the inner and outer membrane. Add the values of the inner and outer membrane to obtain the total lipid molecules per liposome.

•
Convert concentration of lipids to molecules of lipids using Avogadro's number. Using the value determined in the previous step, convert lipid molecules to liposome assemblies.

•
Convert the number of liposome assemblies to molar concentration using Avogadro's number.

NMR Sample Preparation for Liposome Titration
• For all liposome titration experiments, 100 µM protein concentration is used with increasing amounts of liposomes. To achieve the same concentration of protein throughout the titration, a stock of 500 µM of protein is prepared and diluted for each individual sample. CRITICAL STEP: For NMR experiments a final concentration of 10% D2O is needed for frequency locking; therefore, our stock of 500 µM protein is prepared with 50% D2O to ensure that all samples have the proper concentration of both protein and D2O.
• Prepare protein stocks in HPLC-grade water at pH 6.0. Buffer is avoided to allow pH adjustments.

•
Prepare each sample with a volume of 400 µL to compensate for small volume decrease due to pH adjustments. • Table 2 lists values of protein and lipid concentration and volume used in NMR titration experiments. ing the value determined in the previous step, convert lipid molecules to liposome assemblies.

•
Convert the number of liposome assemblies to molar concentration using Avogadro's number.

NMR Sample Preparation for Liposome Titration
• For all liposome titration experiments, 100 µM protein concentration is used with increasing amounts of liposomes. To achieve the same concentration of protein throughout the titration, a stock of 500 µM of protein is prepared and diluted for each individual sample. CRITICAL STEP: For NMR experiments a final concentration of 10% D2O is needed for frequency locking; therefore, our stock of 500 µM protein is prepared with 50% D2O to ensure that all samples have the proper concentration of both protein and D2O.
• Prepare protein stocks in HPLC-grade water at pH 6.0. Buffer is avoided to allow pH adjustments.

•
Prepare each sample with a volume of 400 µL to compensate for small volume decrease due to pH adjustments. • Table 2 lists values of protein and lipid concentration and volume used in NMR titration experiments. • Adjust each sample to pH 6 after mixing.

•
Add 300 µL into a Shigemi tube and insert the plunger avoiding air bubbles at the interface between the sample and glass. Secure plunger and tube with parafilm.

NMR Experiment Set Up for Liposome Titration
NMR experiments using cryogenic probes typically require millimolar or sub-millimolar protein concentrations. In addition, several experiments (approximately 8-10) are performed for NMR titrations. Thus, a significant amount of protein material is necessary at these concentrations, which can turn costly and time-consuming to produce. We recommend decreasing the protein concentration to micromolar values and use fast acquisition techniques [18] to avoid lengthy experiments that could be detrimental for unstable proteins. If SDS-PAGE gel shows positive expression, the glycerol stocks created during this process will be used to provide a stater culture for a larger growth.

Protein Expression in Minimal Media
For NMR studies, it is advised to grow 4 L or more of bacterial culture due to low expression levels.

•
Prepare each 1 L of minimal media in a 6 L flask to provide 5(x) more air for proper aeration of bacterial culture. If SDS-PAGE gel shows positive expression, the glycerol stocks created during this process will be used to provide a stater culture for a larger growth.

Protein Expression in Minimal Media
For NMR studies, it is advised to grow 4 L or more of bacterial culture due to low expression levels.

•
Prepare each 1 L of minimal media in a 6 L flask to provide 5(x) more air for proper aeration of bacterial culture.

Protein Purification
Prepare the reagents for protein resuspension and chromatography as indicated in Section 2. Lightly scrape pellet with 10 mL pipette tip (smaller than the tip of a pencil) and rub pellet onto the walls of a 1.5 mL microcentrifuge tube and wash pellet twice with 200 µL of water, centrifuging in between to remove wash. Resuspend pellet with 40 mL of 1x SDS-PAGE dye.

•
If SDS-PAGE gel revels a significant fraction of the protein remaining in the pellet, resuspend pellet with 20 mL of nickel column buffer A, continue stirring for 30 min and centrifuge again. • Nickel chromatography is performed using a HPLC with a 5 mL prepacked column. CRITICAL STEP: The recommended column flow rate is 1 mL/min. Thus, 5 min is equivalent to 1 column volume (CV).

PAUSE STEP:
The cells can be stored frozen below −20 • C.

Protein Purification
Prepare the reagents for protein resuspension and chromatography as indicated in Section 2. • Nickel chromatography is performed using a HPLC with a 5 mL prepacked column.
ing the value determined in the previous step, convert lipid molecules to liposome assemblies.

•
Convert the number of liposome assemblies to molar concentration using Avogadro's number.

NMR Sample Preparation for Liposome Titration
• For all liposome titration experiments, 100 µM protein concentration is used with increasing amounts of liposomes. To achieve the same concentration of protein throughout the titration, a stock of 500 µM of protein is prepared and diluted for each individual sample. CRITICAL STEP: For NMR experiments a final concentration of 10% D2O is needed for frequency locking; therefore, our stock of 500 µM protein is prepared with 50% D2O to ensure that all samples have the proper concentration of both protein and D2O. • Prepare protein stocks in HPLC-grade water at pH 6.0. Buffer is avoided to allow pH adjustments.

•
Prepare each sample with a volume of 400 µL to compensate for small volume decrease due to pH adjustments. • Table 2 lists values of protein and lipid concentration and volume used in NMR titration experiments. • Adjust each sample to pH 6 after mixing.

•
Add 300 µL into a Shigemi tube and insert the plunger avoiding air bubbles at the interface between the sample and glass. Secure plunger and tube with parafilm.

NMR Experiment Set Up for Liposome Titration
NMR experiments using cryogenic probes typically require millimolar or sub-millimolar protein concentrations. In addition, several experiments (approximately 8-10) are performed for NMR titrations. Thus, a significant amount of protein material is necessary at these concentrations, which can turn costly and time-consuming to produce. We recommend decreasing the protein concentration to micromolar values and use fast acquisition techniques [18] to avoid lengthy experiments that could be detrimental for unstable proteins.
CRITICAL STEP: The recommended column flow rate is 1 mL/min. Thus, 5 min is equivalent to 1 column volume (CV).

•
Purge both nickel column buffers A and B into the system and equilibrate the column by washing with water for 20 min (4 CV), then directly switching to buffer A for an additional 20 min. • An instrument method is created using the time steps shown in Table 1 for the injection and elution of the protein always at a flow rate of 1 mL/min.  where r i is the radius of inner membrane, and r o is the radius of outer membrane both expressed in nm.

•
Determine the surface area (SA) of the inner and outer membrane using Equation (2).
• Divide the surface area of the membrane by the surface area of the head group (0.64 nm 2 ) to determine the total number of lipid molecules in the inner and outer membrane. Add the values of the inner and outer membrane to obtain the total lipid molecules per liposome. • Convert concentration of lipids to molecules of lipids using Avogadro's number. Using the value determined in the previous step, convert lipid molecules to liposome assemblies. • Convert the number of liposome assemblies to molar concentration using Avogadro's number.

NMR Sample Preparation for Liposome Titration
• For all liposome titration experiments, 100 µM protein concentration is used with increasing amounts of liposomes. To achieve the same concentration of protein throughout the titration, a stock of 500 µM of protein is prepared and diluted for each individual sample. • Divide the surface area of the membrane by the surface area of the head group (0.64 nm 2 ) to determine the total number of lipid molecules in the inner and outer membrane. Add the values of the inner and outer membrane to obtain the total lipid molecules per liposome.

•
Convert concentration of lipids to molecules of lipids using Avogadro's number. Using the value determined in the previous step, convert lipid molecules to liposome assemblies.

•
Convert the number of liposome assemblies to molar concentration using Avogadro's number.

NMR Sample Preparation for Liposome Titration
• For all liposome titration experiments, 100 µM protein concentration is used with increasing amounts of liposomes. To achieve the same concentration of protein throughout the titration, a stock of 500 µM of protein is prepared and diluted for each individual sample. CRITICAL STEP: For NMR experiments a final concentration of 10% D2O is needed for frequency locking; therefore, our stock of 500 µM protein is prepared with 50% D2O to ensure that all samples have the proper concentration of both protein and D2O.
• Prepare protein stocks in HPLC-grade water at pH 6.0. Buffer is avoided to allow pH adjustments.

•
Prepare each sample with a volume of 400 µL to compensate for small volume decrease due to pH adjustments. • Table 2 lists values of protein and lipid concentration and volume used in NMR titration experiments. • Adjust each sample to pH 6 after mixing.

•
Add 300 µL into a Shigemi tube and insert the plunger avoiding air bubbles at the interface between the sample and glass. Secure plunger and tube with parafilm.

NMR Experiment Set Up for Liposome Titration
NMR experiments using cryogenic probes typically require millimolar or sub-millimolar protein concentrations. In addition, several experiments (approximately 8-10) are performed for NMR titrations. Thus, a significant amount of protein material is necessary at these concentrations, which can turn costly and time-consuming to produce. We recommend decreasing the protein concentration to micromolar values and use fast acquisition techniques [18] to avoid lengthy experiments that could be detrimental for unstable proteins.

CRITICAL STEP:
For NMR experiments a final concentration of 10% D 2 O is needed for frequency locking; therefore, our stock of 500 µM protein is prepared with 50% D 2 O to ensure that all samples have the proper concentration of both protein and D 2 O. • Prepare protein stocks in HPLC-grade water at pH 6.0. Buffer is avoided to allow pH adjustments.

•
Prepare each sample with a volume of 400 µL to compensate for small volume decrease due to pH adjustments. • Table 2 lists values of protein and lipid concentration and volume used in NMR titration experiments. • Adjust each sample to pH 6 after mixing. • Add 300 µL into a Shigemi tube and insert the plunger avoiding air bubbles at the interface between the sample and glass. Secure plunger and tube with parafilm.

NMR Experiment Set Up for Liposome Titration
NMR experiments using cryogenic probes typically require millimolar or sub-millimolar protein concentrations. In addition, several experiments (approximately 8-10) are performed for NMR titrations. Thus, a significant amount of protein material is necessary at these concentrations, which can turn costly and time-consuming to produce. We recommend decreasing the protein concentration to micromolar values and use fast acquisition techniques [18] to avoid lengthy experiments that could be detrimental for unstable proteins. • Divide the surface area of the membrane by the surface area of the head group (0.64 nm 2 ) to determine the total number of lipid molecules in the inner and outer membrane. Add the values of the inner and outer membrane to obtain the total lipid molecules per liposome.

•
Convert concentration of lipids to molecules of lipids using Avogadro's number. Using the value determined in the previous step, convert lipid molecules to liposome assemblies.

•
Convert the number of liposome assemblies to molar concentration using Avogadro's number.

NMR Sample Preparation for Liposome Titration
• For all liposome titration experiments, 100 µM protein concentration is used with increasing amounts of liposomes. To achieve the same concentration of protein throughout the titration, a stock of 500 µM of protein is prepared and diluted for each individual sample. CRITICAL STEP: For NMR experiments a final concentration of 10% D2O is needed for frequency locking; therefore, our stock of 500 µM protein is prepared with 50% D2O to ensure that all samples have the proper concentration of both protein and D2O.
• Prepare protein stocks in HPLC-grade water at pH 6.0. Buffer is avoided to allow pH adjustments.

•
Prepare each sample with a volume of 400 µL to compensate for small volume decrease due to pH adjustments. • Table 2 lists values of protein and lipid concentration and volume used in NMR titration experiments. • Adjust each sample to pH 6 after mixing.

•
Add 300 µL into a Shigemi tube and insert the plunger avoiding air bubbles at the interface between the sample and glass. Secure plunger and tube with parafilm.

NMR Experiment Set Up for Liposome Titration
NMR experiments using cryogenic probes typically require millimolar or sub-millimolar protein concentrations. In addition, several experiments (approximately 8-10) are performed for NMR titrations. Thus, a significant amount of protein material is necessary at these concentrations, which can turn costly and time-consuming to produce. We recommend decreasing the protein concentration to micromolar values and use fast acquisition techniques [18] to avoid lengthy experiments that could be detrimental for unstable proteins.
CRITICAL STEP: These experiments should be performed in duplicate for error determination. The pKa associated with protein-liposome binding is obtained by fitting the experimental NMR data to the Henderson-Hasselbalch equation (Equation (3)).

NMR Data Processing
• Use Bruker Topsin software for data processing. • Extract the 1D projection of the first serial file for each 2D experiment.

•
Apply phase and baseline correction to the 1D spectrum.

•
Use the first experiment of the titration set (100 µM sapC, no liposomes) as a reference for all subsequent experiments conducted with the same stock solution of protein.

•
By integration, calculate the area within the NMR signals from 9.3 ppm to 6.6 ppm.

•
Normalize the area with respect to the first sample (no liposomes).

•
Determine the integral of all subsequent samples by selecting the option "Use Lastscale as Calibration" from the Topspin software for the same range. • Each normalized integral value indicates the amount of protein remaining in solution; therefore, the relative amount of protein bound to liposomes is calculated with Equation (4).
Relative amount of protein bound to liposomes = 1 − integral, • To determine the K Dapp for the binding between protein and liposome, apply the Hill equation (Equation (5)).

Liposome Fusion
• Liposome fusion was determined using dynamic light scattering (DLS). • Set up parameters for experiment: Refractive index of both protein and liposomes is 1.33, and absorbance at 632.8 nm is 0.05. • Take measurements using the "size" setting with 3 readings per measurement and equilibration time of 120 s.

•
Select the appropriate cuvette. The ZEN0040 cuvette for 160 µL of sample volume is appropriate. • Prepare protein samples for a final protein concentration of 100 µM upon dilution with liposomes. PS liposomes were prepared as described above in Section 3.4 • Filter protein sample through 0.05 µm PES filter and check the initial size and polydispersity prior to liposome addition. • SapC and sapC-PUMA constructs were prepared at pH 4.3 and 6.0, respectively. • Filter liposome sample prior to protein addition to ensure monodispersity. Initial liposome size is determined for time 0. • SapC constructs and liposomes were mixed to achieve a final concentration of 100 µM • SapC and sapC-PUMA constructs were mixed at pH 4.3 and 6.0, respectively. Subsequently, sapC-PUMA solution was adjusted to 5.3.

•
The protein-liposome solution was spun down to pellet any precipitation.

•
The supernatant is then placed into cuvette and tested for liposome fusion. Liposome fusion should immediately begin post centrifugation. For kinetic experiments measurements were taken every 20 min for 200 min.

Expected Results
Saposin C constructs bind to liposomes at acidic or mildly acidic pH. There are three main factors affecting the binding process: solution pH, lipid concentration and type of lipid used to produce liposomes. Binding studies can be performed via titration of the protein with different concentrations of liposomes at a constant pH value or by changing the pH at a constant lipid concentration. In general, it is expected that pH titrations at constant liposome concentration will result in more sapC binding upon further acidification [2]. Analogously, enhanced protein-liposome binding is expected at constant pH and increased liposome concentration [6].
Both titration experiments leverage the significant change in particle size upon binding to liposomes, which slows the effective tumbling rate of the 15 N-labeled protein constructs. Very slow tumbling rates of the proteoliposomes broaden NMR protein signals turning them undistinguishable from the baseline noise level. Under these conditions, only NMR signals from the free protein (unbound) will be observed. This effect is depicted in Figures 3 and 4a, where the NMR signal intensity of the protein sapC-PUMA decreases with the increase in liposome concentration. The decrease in NMR signal intensity indicates protein binding to liposomes (invisible to NMR), thus depleting the solution of free protein (visible to NMR). The change in NMR signal intensity with liposome concentration can be fitted to the Hill equation to determine apparent dissociation constants (K Dapp ) (Figure 4b). The K Dapp found for sapC-PUMA is 3 nM with a Hill coefficient of~2, which indicates the number of adsorbed molecules per binding site in agreement with the dimerization mechanism.
Importantly, this method allows to identify changes in the affinity of protein membrane binding. In fact, the protein sapC-PUMA-DM, with two mutations in the sapC domain of the chimera designed to enhance liposome binding, shows a decreased value of the K Dapp (1.6 nM) [6]. The two mutations replace Asp 52 and Glu 64 in sapC for Arg to reduce the overall negative charge of sapC's electrostatic surface and promote interactions with the negatively charged heads of the phospholipids [2,6].
In addition, NMR titration experiments can also be used to identify the optimal pH for sapC binding to liposomes, resulting in~70% bound protein at pH 4.2 in the presence of PC:PS liposomes [2] (Figure 4c). The fitting of the NMR titration data to the Henderson-Hasselbalch equation allows us to determine an apparent binding pKa of 5.3. The slight deviation from the sigmoidal behavior is likely due to the simultaneous titration of several Glu and Asp side chains in sapC. In fact, the electrostatic surface of sapC is highly negatively charged [2]; the charge is partially neutralized upon acidification, thus facilitating interactions with the hydrophobic chains of the phospholipids.
For certain biotechnological or medical applications of sapC proteoliposomes, it might be necessary to work at mildly acidic conditions. It has been shown that sapC binds preferably to PS lipids as compared to PC or PC/PS mixtures [19]. For this reason, we use PS-only liposomes in the titration experiments to determine the apparent dissociation constant for sapC-PUMA constructs as PS optimizes binding at less acidic conditions (pH 6.0) [6]. We found that the Hill equation is appropriate for determining the K Dapp for liposome binding as it accounts for cooperativity upon binding [20]. SapC and many other saposin-like proteins are known to dimerize [21][22][23][24], suggesting that the Hill coefficient is necessary to account for multiple proteins binding to a single site. the KDapp (1.6 nM) [6]. The two mutations replace Asp 52 and Glu 64 in sapC for Arg to reduce the overall negative charge of sapC's electrostatic surface and promote interactions with the negatively charged heads of the phospholipids [2,6]. In addition, NMR titration experiments can also be used to identify the optimal pH for sapC binding to liposomes, resulting in ~70% bound protein at pH 4.2 in the presence of PC:PS liposomes [2] (Figure 4c). The fitting of the NMR titration data to the Henderson-Hasselbalch equation allows us to determine an apparent binding pKa of 5.3. The slight deviation from the sigmoidal behavior is likely due to the simultaneous titration of several Glu and Asp side chains in sapC. In fact, the electrostatic surface of sapC is highly negatively charged [2]; the charge is partially neutralized upon acidification, thus facilitating interactions with the hydrophobic chains of the phospholipids.  For certain biotechnological or medical applications of sapC proteoliposomes, it might be necessary to work at mildly acidic conditions. It has been shown that sapC binds preferably to PS lipids as compared to PC or PC/PS mixtures [19]. For this reason, we use PS-only liposomes in the titration experiments to determine the apparent dissociation constant for sapC-PUMA constructs as PS optimizes binding at less acidic conditions (pH 6.0) [6]. We found that the Hill equation is appropriate for determining the KDapp for liposome binding as it accounts for cooperativity upon binding [20]. SapC and many other saposinlike proteins are known to dimerize [21][22][23][24], suggesting that the Hill coefficient is necessary to account for multiple proteins binding to a single site.
Some SAPLIPs, such as SP-B, have transient interactions with lipid membranes to facilitate lipid transfer [12]. These transient interactions will likely result in line broadening of NMR signals. Thus, the method as describe here will not be applicable; however, an analogous method that monitors signal line broadening can be devised for such interactions.  Some SAPLIPs, such as SP-B, have transient interactions with lipid membranes to facilitate lipid transfer [12]. These transient interactions will likely result in line broadening of NMR signals. Thus, the method as describe here will not be applicable; however, an analogous method that monitors signal line broadening can be devised for such interactions.
Importantly, the protein purification protocol describes the role of the protease thrombin to remove the His-tag used for purification. This step is necessary because our NMR data indicate that sapC constructs aggregate in the presence of the His-tag at acidic pH ( Figure 5). NMR signals become broader for sapC-PUMA when subjected to pH changes from 6.8 to 4.2 if the His-tag is not removed ( Figure 5). In contrast, this behavior is not observed for sapC or sapC-PUMA constructs without His-tag after pH adjustment from neutral to acidic conditions [2,6]. thrombin to remove the His-tag used for purification. This step is necessary because our NMR data indicate that sapC constructs aggregate in the presence of the His-tag at acidic pH ( Figure 5). NMR signals become broader for sapC-PUMA when subjected to pH changes from 6.8 to 4.2 if the His-tag is not removed ( Figure 5). In contrast, this behavior is not observed for sapC or sapC-PUMA constructs without His-tag after pH adjustment from neutral to acidic conditions [2,6]. In principle, the titrations between SAPLIPs and liposomes can be performed in unlabeled proteins using 1 H 1D-NMR. However, we do not recommend this option because (1) it might be difficult to discern liposome binding from NMR signal broadening, should the protein aggregate; (2) if the presence of liposomes promotes conformational changes in the free protein due to transient protein-lipid interactions, this information could be easily lost in 1D NMR; (3) 1D NMR does not allow to fully discern that the protein conserves the original fold at the different pH values used in the titrations experiments. The use of fast acquisition techniques [18] allows to test protein concentrations in the low micromolar range instead of millimolar concentrations typically used in NMR, which might have important budgetary implications. For the same reason, if liposome-binding proteins tend to self-associate, fast acquisition techniques will facilitate the binding studies as they have been extensively used to monitor protein interactions of self-assembling proteins [25][26][27][28][29].
Finally, it is known that sapC and SAPLIPs can fuse liposomes [5,6,22]. This is another relevant characteristic of the function of SAPLIPs. Fusogenic properties can be leveraged for important biomedical applications involving drug delivery due to the potential fusion of proteoliposomes and cell plasma membranes. DLS is an ideal technique for liposome fusion studies, although fusion can also be assessed with size exclusion chromatography and transmission electron microscopy [19]. In contrast to the latter two, DLS allows to monitor fusion in real time, thus providing an estimate of the time required for fusion to occur, the final polydispersity of the liposome solution and liposome largest size achieved. In principle, the titrations between SAPLIPs and liposomes can be performed in unlabeled proteins using 1 H 1D-NMR. However, we do not recommend this option because (1) it might be difficult to discern liposome binding from NMR signal broadening, should the protein aggregate; (2) if the presence of liposomes promotes conformational changes in the free protein due to transient protein-lipid interactions, this information could be easily lost in 1D NMR; (3) 1D NMR does not allow to fully discern that the protein conserves the original fold at the different pH values used in the titrations experiments. The use of fast acquisition techniques [18] allows to test protein concentrations in the low micromolar range instead of millimolar concentrations typically used in NMR, which might have important budgetary implications. For the same reason, if liposome-binding proteins tend to self-associate, fast acquisition techniques will facilitate the binding studies as they have been extensively used to monitor protein interactions of self-assembling proteins [25][26][27][28][29].
Finally, it is known that sapC and SAPLIPs can fuse liposomes [5,6,22]. This is another relevant characteristic of the function of SAPLIPs. Fusogenic properties can be leveraged for important biomedical applications involving drug delivery due to the potential fusion of proteoliposomes and cell plasma membranes. DLS is an ideal technique for liposome fusion studies, although fusion can also be assessed with size exclusion chromatography and transmission electron microscopy [19]. In contrast to the latter two, DLS allows to monitor fusion in real time, thus providing an estimate of the time required for fusion to occur, the final polydispersity of the liposome solution and liposome largest size achieved.
Prior to fusion, the initial conditions of the liposome sample should be checked, aiming for monodisperse solutions of liposomes with diameters in the 100-200 nm range. Upon addition of sapC constructs, liposome fusion takes place with the most activity occurring in the first five minutes ( Figure 6). In our experiments, liposomes show a nine-fold increase in size over the course of 200 min [6]. No additional changes in liposome size are observed at longer times.
Prior to fusion, the initial conditions of the liposome sample should be checked, aiming for monodisperse solutions of liposomes with diameters in the 100-200 nm range. Upon addition of sapC constructs, liposome fusion takes place with the most activity occurring in the first five minutes ( Figure 6). In our experiments, liposomes show a ninefold increase in size over the course of 200 min [6]. No additional changes in liposome size are observed at longer times.