Deuteration Aiming for Neutron Scattering

Abstract

The distinguished feature of neutron as a scattering probe is an isotope effect, especially the large difference in neutron scattering length between hydrogen and deuterium. The difference renders the different visibility between hydrogenated and deuterated proteins. Therefore, the combination of deuterated protein and neutron scattering enables the selective visualization of a target domain in the complex or a target protein in the multi-component system. Despite of this fascinating character, there exist several problems for the general use of this method: difficulty and high cost for protein deuteration, and control and determination of deuteration ratio of the sample. To resolve them, the protocol of protein deuteration techniques is presented in this report. It is strongly expected that this protocol will offer more opportunity for conducting the neutron scattering studies with deuterated proteins.

Significance

Selective studies on the target protein in the complex or the multi-component system are possible with neutron scattering assisted by deuterated protein. However, following three problems impeded the neutron scattering studies by normal users. The first one is the requirement of deuterated protein of which the deuteration level is controlled. The second one is the difficulty for efficient determination of the degree of deuteration of protein. The third one is the high cost for sample preparation. Then, we provide the protocol of protein deuteration technique, offering more opportunity for neutron scattering coupled with deuterated protein by normal users.

Introduction

The solution scattering techniques are powerful tools for structural and dynamical studies on a large biological complex or system under physiological environments. Especially, neutron solution scattering is quite advanta­geous for the structural and dynamical studies on a large biological complex or system. One of unique features of neutron as a scattering probe is an isotope effect on the scattering length [1], which is particularly evident on hydrogen. To be more specific, the neutron scattering length of hydrogen (H) (–3.739 fm) is markedly different from that of deuterium (D) (6.671 fm), rendering the different visibility to a molecule by replacing the hydrogen with the deuterium. We shall explain the utilization of the isotope effect for the structural study of a large biological complex or system in detail afterwards.

Let’s consider a large biological complex or system comprised of several macromolecular components. These components are protein, nucleic acid, and carbohydrate with different scattering length densities (SLDs) from one another. Here, scattering intensity from one component is proportional to the square of Δρ_component [1], where Δρ_component is called contrast defined as the absolute difference between SLDs of the component (ρ_component) and of solvent (ρ_solvent), Δρ_component=|ρ_component–ρ_solvent|. By setting the condition of Δρ_component=0, namely “matching condition”, the scattering from the component almost becomes disappeared, implying that the component becomes “scatteringly invisible”. This is the most notable technique to selectively observe the target component, as a visible component, in the multi-components system and/or the complex with the solution neutron scattering.

There is one inevitable prerequisite to allow the matching technique: The components in the system should have different SLD values. One of the suitable research targets with this technique is a solution structure of nucleosome, which consists of histones (protein) and DNA. Because the SLD values of histone and DNA are ~2.2×10^–6 Å^–2 and ~3.9×10^–6 Å^–2, respectively, the matching condition for histone is achieved with 40% (v/v) D₂O solvent (ρ_solvent=2.2×10^–6 Å^–2) and that for DNA is also with 65% (v/v) D₂O solvent (ρ_solvent=3.9×10^–6 Å^–2), as shown in Figure 1. Then, we can selectively observe the DNA structure of nucleosome by making the histones scatteringly invisible in 40% (v/v) D₂O solvent, and can selectively observe the histones structure by making the DNA scatteringly invisible in 65% (v/v) D₂O solvent as well. This has been extensively applied for the several structural studies of nucleosomes with small-angle neutron scattering (SANS) [2–4].

Figure 1 

Neutron scattering length densities (SLD) of lipid, hydrogenated protein (h-protein), nucleic acid, partially-deuterated protein (pd-protein) and 100% deuterated protein (100d-protein).

Here, there arises one issue, “How can we selectively observe the target protein when the all components in the complex or system are proteins?” The answer is “protein deuteration”. As shown in Figure 2, the SLD value of the protein is varied with its deuteration ratio. To selectively observe the target subunit in the complex or the target protein in the multi-component solution, the standard combination for the proteins (subunits) is 100% deuterated (100d-) proteins (subunits) and non-deuterated (hydrogenated (h-)) one(s). By using 40% (v/v) D₂O as the solvent, the h-proteins become invisible (refer to Fig. 1) and the 100d-proteins are selectively visible with the high contrast. This method (CM-SANS) using small-angle neutron scattering (SANS) assisted by contrast matching (CM) has solved the structure of the target component in the complex or system [5–7].

Figure 2 

SLD of protein in 0% (red line) and 100% (blue line) D₂O solvent as a function of the degree of deuteration.

However, there is a drawback in CM-SANS method, which is attributed to a large number of hydrogens in 40% (v/v) D₂O solvent. The hydrogen atom causes the intensive incoherent scattering, which becomes the background scattering and deteriorates the signal-to-noise (S/N) ratio, especially in the higher scattering vector (Q) range of SANS intensity profile. To overcome the drawback, inverse contrast matching SANS (iCM-SANS) method has been proposed [8]. In the iCM-SANS, partially deuterated (c.a. 75%) and non-deuterated protein components are ‘invisible’ and ‘visible’, in terms of scattering, in 100% D₂O solvent, respectively. In the iCM-SANS, it should be noted that partially deuterated (pd-) proteins become invisible (refer to Fig. 1) and the h-proteins are selectively observed under the low incoherent background scattering thanks to 100% D₂O as the solvent. Recently, iCM-SANS is utilized for the investigation of partial structures in the complicated complexes [8–10].

As mentioned above, the protein deuteration and related techniques are indispensable for the neutron scattering study with taking the advantage of the neutron characteristics. However, there are obstacles for the normal users to perform the neutron scattering studies with deuterated protein as a conventional technique. In this report, we introduce the full protocol package for CM- and iCM-SANS, and the other neutron solution scattering. This protocol is comprised of following three sections.

In the first section, the procedure for the preparation of deuterated protein of which degree of deuteration is controlled is introduced. In the second section, the efficient procedure for the determination of the degree of deuteration of protein is presented. In the third section, the procedure for recycling of D₂O, which serves to the cost reduction of sample preparation, is summarized. It is strongly expected that this protocol will offer more opportunity for structural study of a large biological complex or system with neutron scattering for the structural biologists from various research fields.

Materials and Methods

Chemical Reagents

Hydrogenated and deuterated D-glucose-1,2,3,4,5,6,6-d7 (97% atom D) were purchased from Wako Pure Chemical Industries (Osaka, Japan) and Isotec Inc. (Miamisburg, OH, USA) respectively. D₂O (99.8% atom D) was purchased from Taiyo Nippon Sanso Corporation (Tokyo, Japan). LC/MS grade 0.1% aqueous trifluoroacetic acid and acetonitrile were purchased from GL Science Inc. (Tokyo, Japan). Sinapinic acid (Matrix substance for MALDI-MS) and protein standard II were purchased from Bruker Daltonics (Billerica, MA, USA). All other chemical reagents were purchased from Sigma Aldrich (St Louis, MO, USA) or Wako Pure Chemical Industries.

MALDI-TOF Mass Spectrometry (MALDI-TOF MS)

The protein sample solution and a saturated solution of sinapinic acid in TA30 (30% acetonitrile in 70% of 0.1% TFA in water) were mixed with the volume ratio of 1 to 9 and dropped on the ground steel MALDI plate: 1 μl protein for sample solution and 9 μl for saturated sinapinic acid one in the standard case. After drying of MALDI-TOF MS sample and crystallization of droplets, the measurement was conducted with a microflexLT MALDI TOF Mass Spectrometer (Bruker Daltonics) in positive ion mode. External calibrations were performed using protein standards II. Mass spectrum raw data was recorded with flexControl and analyzed with flexAnalysis (Bruker Daltonics).

Small-angle neutron scattering (SANS)

SANS experiments were performed with Quokka [11] installed at the Australian Nuclear Science and Technology Organization (ANSTO, Lucas Heights, NSW, Australia). The wavelength and the sample-to-camera distance were 6.0 Å and 6.0 m, respectively, covering the Q range from 0.012 Å^–1 to 0.10 Å^–1. The obtained SANS intensity profiles were corrected for background, empty cell, buffer solution scattering, and transmittance. Utilizing the incident beam flux, the SANS intensity profiles were converted to absolute intensities (cm⁻¹).

Fourier transform infrared (FT-IR) spectroscopy

The IR spectra were measured with FT/IR-4600 (Jasco, Tokyo, Japan) equipped with a triglycine sulfate detector, Ge/KBr beam splitter and ATR PRO ONE. The sample solution with the volume of 3.0 μL was placed on the center of diamond prism and immediately measured. Spectra were recorded in the wavenumber covering from 500 to 4000 cm^–1 with the resolution of 4 cm^–1 at a scan rate of 2 mm/min. To secure high S/N ratio of observed spectrum, the spectrum was finally acquired after averaging of spectra from 32 scans using the Spectra Manager (Jasco).

Evaporator

To collect the D₂O from the used 100% D₂O containing M9 medium (used 100% D₂O M9 medium), used 100% D₂O M9 medium was evaporated with the Rotavapor^® R-300 (Buchi, Flawil, Switzerland) equipped with the vacuum pump V-300 (Buchi), the heating bath B-305 (Buchi) and the circulating water pump LTB-125 (AS ONE Co., Osaka, Japan). The circulating water of which temperature was set to 4°C, was connected to the reflux tube of the evaporator. Flask filled with 100% D₂O M9 medium was immersed in the heating bath at 42°C and rotated at 140 rpm. The internal pressure was initially set at 200 mbar and then reduced to 100 mbar step-wisely. After confirming the absence of sudden boiling, the internal pressure was gradually reduced to 30 mbar.

Software

VolumeCalc1 calculates the numbers of exchangeable and non-exchangeable H atoms with a sequence of one- or three-letter abbreviation of amino acid residues as an input file (ASCI file). For the calculation, VolumeCalc1 obtained the number of non-exchangeable hydrogen atoms in each amino acid residue from the report of Jacrot [12]. In addition, VolumeCalc1 also provides for the expected SLD of a concerned protein corresponding to the degree of deuteration of protein and the volume fraction of D₂O in the buffer.

D₂O/H₂O ratio calculator (d2or) is a Python script using the Numpy library. The peaks at around 1200 cm^–1 (Peak 1) and 1450 cm^–1 (Peak 2) [13] correspond to the frequency of bending vibration of D-O-D bonding and that of H-O-D bonding, respectively (Fig. 3 (a)). By changing the volume fraction of D₂O, we measured IR spectra and calculated the integrated intensity from 1150 cm^–1 to 1250 cm^–1 (area 1) and that from 1400 cm^–1 to 1500 cm^–1 (area 2). From the ratio of area 2 to area 1 as a function of volume fraction of D₂O, the calibration curve was determined (Fig. 3 (b)). With this calibration curve, the D₂O/H₂O ratio of a concerned buffer can be computed by supplying the ratio of area 2 to area 1 as an input parameter.

Figure 3 

(a) IR spectra from D₂O/H₂O ratio of 30.0% (v/v) (orange line), 40.0% (v/v) (lemon line), 50.0% (v/v) (light green line), 60.0% (v/v) (green line), 70.0% (v/v) (cyan line), 80.0% (v/v) (light blue line), 90.0% (v/v) (blue line) and 99.96% (v/v) (purple line), respectively. (b) Area ratio of Peak 2 to Peak 1 as a function of D₂O/H₂O ratio (circle) and solid line corresponds to fit with a polynomial equation.

Software is available from following our website (http://www.rri.kyoto-u.ac.jp/NSBNG/activity.html).

Protocols

Preparation of 100% and partially deuterated protein samples

Purpose

As described in the introduction, contrast Δρ_component is a crucial parameter that determines “scatteringly visibility” (or “scatteringly invisibility”) of solute in the solution. The forward scattering intensity (I₀) is related to Δρ_component through the following equation.

  

$$I_0={NV}^2{({\rho }_{\mathrm{component}}–{\rho }_{\mathrm{solvent}})}^2={NV}^2{{\Delta \rho }_{\mathrm{component}}}^2\text{,}$$eq. (1)

where N and V correspond to the number density and volume of component (solute), respectively. Referring to Figure 1, the SLDs of h- and pd-proteins are scatteringly invisible in 40% (v/v) D₂O solvent and 100% D₂O solvent, respectively. Taking advantage of their properties, the combination of hydrogenated (h-) and 100%-deuterated (100d-) proteins is utilized for CM-SANS. And the combination of h- and pd-proteins is utilized for iCM-SANS.

In this section, we introduce the procedures for the preparation of recombinant 100d- and pd-proteins with the E. coli expression system.

Preparations for pre-culture and culture of E. coli for 100d- and pd-proteins

Prior to the pre-culture of E. coli for 100d-protein, the following procedures are needed.

0. H₂O should be sterilized with autoclaving.

1. Referring to the ingredients summarized in Table 1, 5 times concentrated LB broth (5× LB broth) is prepared. This is utilized for the pre-culture of both 100d- and pd-proteins.

Table 1  Ingredients of 5× LB broth

5× LB broth		1× concentration
bacto-tryptone	50.0 g	10.0 g/L
bacto-yeast extract	25.0 g	5.0 g/L
NaCl	50.0 g	10.0 g/L
NaOH		adjust to pH 7.5
D. W	up to 1 L
Total	1 L	*autoclave

2. Referring to the ingredients in Table 2, 10 times concentrated M9 medium (10× M9 medium) is prepared. This is used for the culture of pd-protein.

Table 2  Ingredients of 10× M9 medium

10× M9 medium		1× concentration
NH4Cl	10.0 g	18.7 mM
KH2PO4	20.0 g	14.7 mM
Na2HPO4	63.4 g	44.7 mM
NaCl	5.0 g	8.6 mM
H2O	up to 1 L
Total	1 L	*autoclave

3. Referring to the ingredients in Table 3, the additive solution for M9 medium with the volume of 1 L is prepared. To avoid the contamination of H₂O to M9 medium, this solution is applied to freeze-drying. After freeze-drying, it is stored at –20 °C.

Table 3  Ingredients of additive solution for 1 L M9 medium

Additive solution for 1 L M9 medium		final concentration
2 M MgSO4	1.0 mL	2.0 mM
1 M CaCl2	100.0 μL	100.0 μM
0.1 M FeCl3	216.0 μL	21.6 μM
50 mg/mL Thiamin	48.0 μL	2.4 mg/L
1 mg/mL Biotin	1.2 mL	1.2 mg/L
Total	2.6 mL	*freeze-dry

4. The solutions of appropriate inducer (ex. isopropyl β-D-thiogalactopyranoside) and antibiotic (ex. Ampicillin) are prepared in H₂O. These solutions are sterilized with a filter having the pore size of 0.22 μm. To avoid the contamination of H₂O to M9 medium, the filtered solutions are applied to freeze-drying. After freeze-drying, they are stored at –20°C.

5. Flask for the culture, the filtration device, the culture tubes and so on are autoclaved. They are dried at 100°C for at least 4 days to avoid the contamination of H₂O.

Culture for 100d-protein

Firstly, the adaptation of E. coli to the D₂O environment is indispensable for avoiding long, and variable lag time of growth [14]. The volume fraction of D₂O in the culture solution must be step-wisely increased at the stage of pre-culture, hence four steps are needed for the culture for 100d-protein. In the following, we briefly summarize the procedures at each step and further detailed procedures are shown in Figure 4.

Figure 4 

The schematic procedures for the cultivation of 100d-protein.

Step 1: 1st pre-culture

Referring to the ingredients in Table 4, LB broth containing 30% D₂O (30% D₂O LB broth) with the volume of 5 mL is prepared. Then, 30% D₂O LB broth is inoculated from the glycerol stock solution of E. coli. The bacterial cells are cultured in 30% D₂O LB broth for over 12 h at 37°C.

Table 4  Ingredients of 30%, 60% and 80% D₂O LB broth

30% D2O LB broth
5× LB broth	1.0 mL
filter-sterilized D2O	1.5 mL
sterile H2O	2.5 mL
antibiotic (in H2O)
Total	5.0 mL
60% D2O LB broth
5× LB broth	1.0 mL
filter-sterilized D2O	3.0 mL
sterile H2O	1.0 mL
antibiotic (in H2O)
Total	5.0 mL
80% D2O LB broth
5× LB broth	1.0 mL
filter-sterilized D2O	4.0 mL
antibiotic (in H2O)
Total	5.0 mL

Step 2: 2nd pre-culture

Referring to the ingredients in Table 4, LB broth containing 60% D₂O (60% D₂O LB broth) with the volume of 5 mL is prepared. Then, 50 μL of turbid 30% D₂O LB broth is added to 60% D₂O LB broth. E. coli is cultured in 60% D₂O LB broth for over 12 h at 37°C.

Step 3: 3rd pre-culture

Referring to the ingredients in Table 4, LB broth containing 80% D₂O (80% D₂O LB broth) with the volume of 10 mL is prepared. Then, 40 μL of turbid 60% D₂O LB broth is added to 80% D₂O LB broth. E. coli is cultured in 80% D₂O LB broth for over 12 h at 37°C.

Step 4: Main culture

Referring to the ingredients in Table 5, 100% D₂O M9 medium with the volume of 1 L is prepared for the main culture. After dissolving the additive of M9 medium and the antibiotic in 100% D₂O M9 medium, 100% D₂O M9 medium is applied to the sterilized-filter. The cells in 80% D₂O LB culture broth are collected by centrifugation and the collected cells are suspended against 100% D₂O M9 medium. Then, the suspended solution of the cells is cultured in 100% D₂O M9 medium at 37°C until the optical density of 600 nm (OD₆₀₀) reaching 0.6~1.0. Finally, the protein expression is induced at the appropriate condition after adding the inducer.

Table 5  Ingredients of 100% D₂O M9 medium

Ingredient	weight/volume	final concentration
NH4Cl	1.0 g	18.7 mM
KH2PO4	2.0 g	14.7 mM
Na2HPO4	6.34 g	44.7 mM
NaCl	0.5 g	8.6 mM
deuterated D-glucose	2.0 g	2.0 mg/mL
D2O	1 L
freeze-dried antibiotic	appropriate weight	appropriate concentration
freeze-dried additive	Refer to Table 3
Total	1 L	*filter sterilize

Culture for pd-protein

Three steps are needed for the culture for pd-protein. In the following, we briefly summarize the procedures at each step and further detailed procedures are presented in Figure 5.

Figure 5 

The schematic procedures for the cultivation of pd-protein.

Step 1: 1st pre-culture

Referring to the ingredients in Table 4, 30% D₂O LB broth with the volume of 5 mL is prepared. Then, 30% D₂O LB broth is inoculated from the glycerol stock solution of E. coli. The bacterial cells are cultured in 30% D₂O LB broth for over 12 h at 37°C.

Step 2: 2nd pre-culture

Referring to the ingredients in Table 4, 60% D₂O LB broth with the volume of 10 mL is prepared. Then, 100 μL of turbid 30% D₂O LB broth is added to 60% D₂O LB broth. E. coli is cultured in 60% D₂O LB broth for over 12 h at 37°C.

Step 3: Main culture

Referring to the ingredients in Table 6, 75% D₂O M9 medium with the volume of 1 L is prepared for the main culture. After dissolving the additive of M9 medium and the antibiotic in 75% D₂O M9 medium, 75% D₂O M9 medium is applied to the sterilized-filter. The cells in 60% D₂O LB culture broth are collected by centrifugation and the collected cells are suspended against 75% D₂O M9 medium. Then, the suspended solution of the cells is cultured in 75% D₂O M9 medium at 37°C until OD₆₀₀ reaching 0.6~1.0. Finally, the protein expression is induced at the appropriate condition after adding the inducer.

Table 6  Ingredients of 75% D₂O M9 medium

Ingredient	weight/volume	final concentration
10× M9 medium	100.0 mL
deuterated D-glucose	1.5 g	1.5 mg/mL
hydrogenated D-glucose	0.5 g	0.5 mg/mL
D2O	750.0 mL
H2O	150.0 mL
freeze-dried antibiotic	appropriate weight	appropriate concentration
freeze-dried additive	Refer to Table 3
Total	1 L	*filter sterilize

Following the procedures of culture for 100d-protein, deuterated synechococcus KaiA (SyKaiA) was obtained with an amount of ca. 5 mg after appropriate purification steps. Following the procedures of culture for pd-protein, pd-αB-crystallin (αB-cry) was obtained with an amount of ca. 6 mg after appropriate purification steps.

Efficient determination of the degree of deuteration of protein and D₂O/H₂O ratio in buffer

Purpose

As given in eq. (1), both ρ_component and ρ_solvent affect the observable I₀ value. To realize a successful iCM (or CM) SANS study, both the degree deuteration of protein and the D₂O/H₂O ratio in the buffer should be determined accurately. In this section, we firstly introduce the procedure that can efficiently determine the degree of deuteration of protein through MALDI-TOF MS. As a next step, we also introduce the procedure that can efficiently determine the D₂O/H₂O ratio in the buffer through FT-IR spectroscopy.

Efficient determination of the degree of deuteration of protein with MALDI-TOF MS

Because the nonvolatile ingredients in the buffer such as Tris and NaCl sometimes retards the ionization of a sample, the desalting of a sample is normally required prior to MALDI-TOF MS measurement. On the other hand, we have succeeded to obtain a mass spectrum with high S/N even from the non-desalting sample with the following two technical tips. The first one is the adjustment of the molar ratio of a protein sample solution and saturated sinapinic acid one of 1 to 1000 in the MALDI-TOF MS sample. The second one is the preparation of a protein sample solution at the concentration of 5.0~10.0 mg/mL. Considering the above-described experimental requirements, SyKaiA solu­tion at the concentration of 6.0 mg/mL and αB-cry at the concentration of 8.0 mg/mL were utilized for the preparation of MALDI-TOF MS samples.

The buffer containing D₂O is normally utilized for neutron scattering, hence it is desirable for determining the degree of deuteration of protein in the buffer used for neutron scattering. On the other hand, it was reported that the exchange rates from H atoms to D atoms of the exchangeable H atoms of amino acid residues in D₂O were different, reflecting the variation of the local environment of amino acid residues in the protein [15]. It makes the coexistence of H and D atoms in the exchangeable H atoms of protein, hindering the accurate determination of mass of 100d- (or pd-) protein. As a result, the degree of deuteration of 100d- (or pd-) protein cannot be determined accurately from a single mass spectrum of 100d- (or pd-) protein. To overcome such a problematic situation, we have to consider an alternative procedure.

Since the number of non-exchangeable H atoms in the protein is not affected by the D₂O/H₂O ratio in the buffer, the number of non-exchangeable D atoms in the deuterated protein is usable for the determination of the degree of deuteration of protein. We then adopt the following procedures.

1. Both deuterated and hydrogenated proteins must be dialyzed against the buffer with the same D₂O/H₂O ratio for the same length of time.

2. MALDI-TOF MS measurements are performed for both samples.

3. Because the peak shift between 100d- (or pd-) and h-protein in the MS spectra corresponds to the number of non-exchangeable D atoms in the deuterated protein, the peak shift value between two samples (Δm/z) is calculated.

4. The degree of deuteration of protein can be calculated from dividing Δm/z by the total number of non-exchangeable H atoms in the h-protein.

Figure 6 shows the MS spectra of hydrogenated and deuterated SyKaiA and those of hydrogenated and partially deuterated αB-cry. There are several peaks in the mass spectra. We adopted the peaks with the lowest m/z (indicated by solid arrows) for the calculation of the degree of deuteration of protein because they have less effect from the solvents. In the present work, the averaged Δm/z values from three different sample lots were utilized for further analysis. The numbers of total non-exchangeable H atoms in h-protein and Δm/z values of SyKaiA and αB-cry are summarized in Table 7 and the degrees of deuteration of SyKaiA and αB-cry are calculated to 101.0±1.5% and 70.4±1.2%, respectively. It implies that the expected degrees of deuteration are approximately realized by the above-described cultivation procedures. It should be stressed that the degree of deuteration of protein can be determined within one hour from our protocol.

Figure 6 

Mass spectra of hydrogenated (a) and deuterated (b) SyKaiA. Mass spectra of hydrogenated (c) and deuterated (d) αB-cry.

Table 7  Numbers of non-exchangeable H atoms in h-protein, Δm/z values and the degrees of deuteration of SyKaiA and αB-cry

	number of non-exchangeable H atoms in h-protein	Δm/z	degree of deuteration (%)
SyKaiA	1747	1764.5±26.2	101.0±1.5
αB-cry	1093	773.2±13.0	70.4±1.2

To check whether this 70.4% deuterated αB-cry is “scatteringly invisible” in 100% D₂O solvent or not, SANS measurements were performed with this αB-cry in 0%, 40% and 100% D₂O solvent (Fig. 7 (a)). With increasing the volume fraction of D₂O, the scattering intensity decreased and 70.4% deuterated αB-cry was almost scatteringly invisible in 100% D₂O. With this data, we also plotted (I₀/c)^0.5, which is comparable to Δρ_component in eq. (1), as a function of volume fraction of D₂O (Fig. 7 (b)). It was revealed that (I₀/c)^0.5 was equal to 0 at the volume fraction of 98.8±1.5% D₂O, supporting its scattering invisibility in 100% D₂O buffer.

Figure 7 

(a) SANS intensity profiles as a function of Q of pd-αB-cry in 0% (blue circle), 40% (v/v) (red triangle) or 100% (green square) D₂O solvent. (b) (I₀/c)^0.5 as a function of volume fraction of D₂O. The solid line corresponds to the result of fit with a linear function.

To assess the tolerance of deviation from the exact degree of deuteration that can be matched in 100% D₂O solvent (exact degree of deuteration point), we also calculated the expected I₀ value as a function of the degree of deuteration of αB-cry in 100% D₂O. As shown by the solid arrow in Figure 8, it was revealed that 73.2% deuterated αB-cry is exactly scatteringly invisible in 100% D₂O solvent. Assuming the case that the degree of deuteration of pd-αB-cry was 5% deviated from 73.2%, the expected I₀ value of 68.2% (or 78.2%) deuterated αB-cry is computed to 2.5×10^–3 cm^–1. This value approxi­mately corresponds to 1/200 of I₀ value of h-αB-cry in 100% D₂O solvent, implying the fulfillment of the condi­tion of scattering invisibility. Hence, it is considered that less than 5% deviation from the exact degree of deuteration point is demanded for conducting iCM-SANS study.

Figure 8 

The calculated I₀ value as a function of the degree of deuteration level of αB-cry in 100% D₂O. Inset indicates its magnified plot at around the exact degree of deuteration point (=73.2%), indicated by the solid arrow.

Efficient determination of D₂O/H₂O ratio in the buffer

Prior to neutron scattering, the sample is normally dialyzed against buffer including D₂O. During the dialysis process, it is not possible to avoid the contamination of H₂O from the ambient air absolutely. It prohibits the preparation of strictly controlled D₂O/H₂O ratio in the buffer, hence we also have to grasp the real D₂O/H₂O ratio in the buffer for neutron scattering. Here, we also introduce the efficient method for determining the D₂O/H₂O ratio in the buffer by FT-IR spectroscopy. As we have already explained in Materials and Methods, the ratio of the peak area at around 1450 cm^–1 (Peak 2) to that of 1200 cm^–1 (Peak 1) is usable for the determination of D₂O/H₂O ratio in the buffer.

To inspect whether the salt in the buffer affects the IR spectrum or not, we prepared D₂O buffer comprising of 100 mM Tris/HCl (pH 7.4/pD 7.0), 150 mM NaCl and 90% D₂O. The IR spectra from this 90% D₂O (blue line), 99.96% D₂O (green line) and D₂O buffer (red line) are shown in Figure 9. The D₂O/H₂O ratio of the buffer was calculated to 89.7±0.5%, meaning that the D₂O/H₂O ratio is not affected by the salt in the buffer. Hence, D₂O/H₂O ratio of the buffer can be correctly determined through our procedure. With this technique, the D₂O/H₂O ratio in the buffer can be determined within five minutes.

Figure 9 

IR spectra of 90.0% (v/v) D₂O (blue line), 99.96% (v/v) D₂O (green line) and D₂O buffer (red line).

Recycling of D₂O

Purpose

Depending upon the measurement type, the sample volume ranging from 280 μL to 6.0 mL is normally necessary for conducting the neutron scattering. It also means that a large amount of expensive deuterated chemical reagents are required for the sample preparation. Then, we also have to consider the method for reducing the cost for sample preparation. Since 100% D₂O is not necessary for the preparation of pd-protein, recycling of D₂O from D₂O containing M9 medium would contribute to the reduction of the cost for the preparation of pd-protein. In this section, we introduce the procedures for recycling of D₂O from the used 100% D₂O M9 medium with an evaporator.

Procedure for recycling of D₂O

To remove the contaminants from the used 100% D₂O M9 medium, the used medium is firstly centrifuged at 7,500×g for 15 min over prior to the commencement of evaporation. After the centrifuge, the supernatant is trans­ferred to a flask of an evaporator. For efficient evaporation, the volume of the supernatant should not exceed 60% of the volume of the flask. Following the procedure described in Materials and Methods, the evaporation of used 100% D₂O M9 medium was commenced. Distillation of D₂O solvent with the volume of 300 mL was completed within 1–2 h. After distillation, a viscous residual material remain at the bottom of the flask (Fig. 10 (a)). Then, the recycled D₂O is transferred from the receiver flask to a dry medium bottle. By adding 1–2 dispensing spoons of activated charcoal to the medium bottle, the bottle cap is tightly sealed with Parafilm (Fig. 10 (b)). Just prior to reuse, activated charcoal is removed from the recycled D₂O by filtering with the pore size of 0.22 μm. If there is no loss during the evaporation process, about 95% of the initial volume could be recollected. The D₂O/H₂O ratio of recycled D₂O was calculated to be 99.5±0.2% (Fig. 10 (c)), supporting the successful recycling of used D₂O from our protocol.

Figure 10 

(a) Viscous residual material left after evaporation. (b) Recycled D₂O with activated charcoal in the medium bottle. (c) IR spectra of 99.96% (v/v) D₂O (purple line) and recycled D₂O (pink line) from 100% D₂O M9 medium.

Considering both the D₂O/H₂O ratio and collected volume of recycled D₂O, it is possible to realize approxi­mately 50% cost reduction for the preparation of pd-protein. Compared to the required sample volume for SANS (280~560 μL), the required sample volume for quasielastic neutron scattering (>1 mL) is still a big load for normal users. Hence, the cost reduction of sample preparation of deuterated protein will offer more opportunity for studying the dynamics of the target domain in the complex or target protein in the multi-component system with quasielastic neutron scattering.

Future perspective

Recently, more elaborate protein deuteration techniques such as selective isotopic labeling of specific amino acid residue [16], specific region [17] and specific domain in multi-domain [18] are under progress to realize the observation of functionally relevant structure and dynamics of specific sites in the protein. As a next research target, we also have started to prepare the selectively deuterated domain in the multi-domain protein assisted by segment ligation technique [19]. When the efficient segment ligation technique coupled with deuteration one will be established, this protocol will be reported in near future.

Conclusion

The combination of deuterated protein and neutron scattering enables to selectively visualize the target domain in the protein complex or the target protein in the multi-component system. Despite of its unique and notable property, following three obstacles mainly hampered the normal users to conduct the neutron scattering studies on deuterated protein. The first one is the requirement of the deuterated sample of which the degree of deuteration is controlled. The second one is the difficulty for the efficient determination of the degree of deuteration of the sample. The third one is the requirement of a large amount of expensive deuterated reagents. To overcome the above-described problems, the protocol of protein deuteration technique was presented in this report. In the first section, the procedures for the preparation of 100d- and pd-protein were presented. In the second section, the procedures concerning the efficient determination of the degree of deuteration of protein and the D₂O/H₂O ratio in the buffer were introduced. In the third section, the procedure for recycling of D₂O from the used 100% D₂O M9 medium was summarized.

It is strongly expected that this basic protocol will offer more opportunity to perform the neutron scattering studies on deuterated protein by normal users.

Acknowledgement

This work was supported by MEXT/JSPS KAKENHI Grant Numbers (JP19KK0071, JP20K06579 to R. I., JP19K16088 to K. M., JP20K22629 to M. Shimizu, JP17K07816 to N. S., JP18H05229, JP18H05534, JP18H03681 to M. Sugiyama) and by the JSPS Research Fellowship for Young Scientists (to Y. Y.). This was also partially supported by the Sasakawa Scientific Research Grant from The Japan Science Society to A. O, by the visiting Researcher’s Program of the Institute for Integrated Radiation and Nuclear Science, Kyoto University to M. Y.-U. and H. Y., by Joint Research of the Exploratory Research Center on Life and Living Systems (ExCELLS) (ExCELLS program No. 20-326 to M.Sugiyama and No. 20-308 to H.Y.), and by Joint Research by Institute for Molecular Science (IMS program No. 19-205 and 20-207 to M.Sugiyama). The SANS experiments using Quokka at ANSTO were performed under Proposal No. 5744 and 7005 and we are grateful for Dr. Kathleen Wood for Quokka measurement. This study was partially supported by the Radiation Application Development Association (RADA) under the MEXT Support Program for Overseas Facility Users. This work was also partially supported by KumatorI Deuteration Station (KIDS) Project and the project for construction of the basis for advanced materials science and analytical study by the innovative use of quantum beam and nuclear sciences at the Institute for Integrated Radiation and Nuclear Science, Kyoto University.

Conflicts of Interest

The authors declare no competing interests.

Author Contributions

A. O., R. I., Y. Y., M. Y. -U., H. Y., M. Shimizu, R. U., K. K., and M. Sugiyama contributed to the establishment of the protocol for the preparation of 100d- and pd-protein. A. O., and N. S. performed MALDI-TOF MS measurements on 100d-, pd- and h-protein and analyzed the mass spectra. R. I., and M. Sugiyama performed SANS measurements on pd-αB-cry solutions and analyzed SANS intensity profiles. K. M. performed FT-IR spectroscopy measurements on the D₂O buffer and analyzed the spectra. A. O., R. I., T. S., and N. S. contributed to the establishment of the protocol for recycling of D₂O with an evaporator. M. Sugiyama and R. I. created the program of “VolumeCalc1” and “d2or”. R. I., and M. Sugiyama designed the research and all of the authors wrote the paper.

References

• [1]  Fitter, J., Gutberlet, T. & Katsaras, J. Neutron Scattering in Biology, Techniques and Applications (Springer, Heidelberg, 2006).
• [2]   Hjelm,  R. P.,  Kneale,  G. G.,  Sauau,  P.,  Baldwin,  J. P.,  Bradbury,  E. M. &  Ibel,  K. Small angle neutron scattering studies of chromatin subunits in solution. Cell 10, 139–151 (1977). DOI: 10.1016/0092-8674(77)90148-9
• [3]   Sugiyama,  M.,  Arimura,  Y.,  Shirayama,  K.,  Fujita,  R.,  Oba,  Y.,  Sato,  N., et al. Distinct features of the histone core structure in nucleosomes containing the histone H2A.B variant. Biophys. J. 106, 2206–2213 (2014). DOI: 10.1016/j.bpj.2014.04.007
• [4]   Matsumoto,  A.,  Sugiyama,  S.,  Li,  Z.,  Martel,  A.,  Porcar,  L.,  Inoue,  R., et al. Structural studies of overlapping dinucleo­somes in solution. Biophys. J. 118, 2209–2219 (2020). DOI: 10.1016/j.bpj.2019.12.010
• [5]   Capel,  M. S.,  Engelman,  D. M.,  Freeborn,  B. R.,  Kjeldgaard,  M.,  Langer,  J. A.,  Ramakrishnan,  V., et al. A Complete mapping of the proteins in the small ribosomal subunit of Escherichia coli. Science 238, 1403–1406 (1987). DOI: 10.1126/science.3317832
• [6]   Johansen,  D.,  Jeffries,  C. M. J.,  Hammouda,  B.,  Trewhella,  J. &  Goldenberg,  D. P. Effects of macromolecular crowding on an intrinsically disordered protein characterized by small-angle neutron scattering with contrast matching. Biophys. J. 100, 1120–1128 (2011). DOI: 10.1016/j.bpj.2011.01.020
• [7]   Ibrahim,  Z.,  Martel,  A.,  Moulin,  M.,  Kim,  H. S.,  Härtlein,  M.,  Franzetti,  B., et al. Time-resolved neutron scattering provides new insight into protein substrate processing by a AAA+ unfoldase. Sci. Rep. 7, 40948 (2017). DOI: 10.1038/srep40948
• [8]   Sugiyama,  M.,  Yagi,  H.,  Yamaguchi,  T.,  Kumoi,  K.,  Hirai,  M.,  Oba,  Y., et al. Conformational characterization of a protein complex involving intrinsically disordered protein by small-angle neutron scattering using the inverse contrast matching method: a case study of interaction between a-synuclein and PbaB tetramer as a model chaperone. J. Appl. Cryst. 47, 430–435 (2014). DOI: 10.1107/S1600576713033475
• [9]   Sugiyama,  M.,  Yagi,  H.,  Ishii,  K.,  Porcar,  L.,  Martel,  A.,  Oyama,  K., et al. Structural characterization of the circadian clock protein complex composed of KaiB and KaiC by inverse contrast-matching small-angle neutron scattering. Sci. Rep. 6, 35567 (2016). DOI: 10.1038/srep35567
• [10]   Yagi-Utumi,  M.,  Sikdar,  A.,  Song,  C.,  Park,  J.,  Inoue,  R.,  Watanabe,  H., et al. Supramolecular tholos-like architecture constituted by archaeal proteins without functional annotation. Sci. Rep. 10, 1540 (2020). DOI: 10.1038/s41598-020-58371-2
• [11]   Wood,  K.,  Mata,  J. P.,  Garvey,  C. J.,  Wu,  C.-M.,  Hamilton,  W. A.,  Abbeywick,  P., et al. QUOKKA, the pinhole small-angle neutron scattering instrument at the OPAL Research Reactor, Australia: design, performance, operation and scientific highlights. J. Appl. Cryst. 51, 294–314 (2018). DOI: 10.1107/S1600576718002534
• [12]   Jacrot,  B. The study of biological structures by neutron scattering from solution. Rep. Prog. Phys. 39, 911–953 (1976). DOI: 10.1088/0034-4885/39/10/001
• [13]   Venyaminov,  S. Y. &  Prendergast,  F. G. Water (H₂O and D₂O) molar absorptivity in the 1000–4000 cm^–1 range and quantitative infrared spectroscopy of aqueous solutions. Anal. Biochem. 248, 234–245 (1997). DOI: 10.1006/abio.1997.2136
• [14]   Duff,  A. P.,  Wilde,  K. L.,  Rekas,  A.,  Lake,  V. &  Holden,  P. J. Robust high-yield methodologies for ²H and ²H/¹⁵N/¹³C labeling of proteins for structural investigations using neutron scattering and NMR. Methods Enzymol. 565, 3–25 (2015). DOI: 10.1016/bs.mie.2015.06.014
• [15]   Kostyukevich,  Y.,  Acter,  T.,  Zherebker,  A.,  Ahmed,  A.,  Kim,  S. &  Nikolaev,  E. Hydrogen/deuterium exchange in mass spectrometry. Mass Spectrom. Rev. 37, 811–853 (2018). DOI: 10.1002/mas.21565.
• [16]   Wood,  K.,  Gallat,  F.-X.,  Otten,  R.,  van Heel,  A. J.,  Lethier,  M.,  van Eijck,  L., et al. Protein surface and core dynamics show concerted hydration-dependent activation. Angew. Chem., Int. Ed. Engl. 52, 665–668 (2013). DOI: 10.1002/anie.201205898
• [17]   Nickels,  J. D.,  Cheng,  X.,  Mostofian,  B.,  Stanley,  C.,  Lindner,  B.,  Heberle,  F. A., et al. Mechanical properties of nanoscopic lipid domains. J. Am. Chem. Soc. 137, 15772–15780 (2015). DOI: 10.1021/jacs.5b08894
• [18]   Sonntag,  M.,  Jagtap,  P. K. A.,  Simon,  B.,  Appavou,  M.-S.,  Geerlof,  A.,  Stehle,  R., et al. Segmental, domain-selective perdeuteration and small-angle neutron scattering for structural analysis of multi-domain proteins. Angew. Chem. Int. Ed. Engl. 56, 9322–9325 (2017). DOI: 10.1002/anie.201702904
• [19]   Freiburger,  L.,  Sonntag,  M.,  Hennig,  J.,  Li,  J.,  Zou,  P. &  Sattler,  M. Efficient segmental isotope labeling of multi-domain proteins using Sortase A. J. Biomol. NMR 63, 1–8 (2015). DOI: 10.1007/s10858-015-9981-0