Journal of Magnetic Resonance 257 (2015) 32–38

Contents lists available at ScienceDirect

Journal of Magnetic Resonance journal homepage: www.elsevier.com/locate/jmr

Dynamic nuclear polarization properties of nitroxyl radical in high viscous liquid using Overhauser-enhanced Magnetic Resonance Imaging (OMRI) M. Kumara Dhas a, Hideo Utsumi b, A. Jawahar c, A. Milton Franklin Benial a,⇑ a b c

Department of Physics, NMSSVN College, Nagamalai, Madurai 625 019, Tamil Nadu, India Innovation Center for Medical Redox Navigation, Kyushu University, Fukuoka 812-8582, Japan Department of Chemistry, NMSSVN College, Nagamalai, Madurai 625 019, Tamil Nadu, India

a r t i c l e

i n f o

Article history: Received 3 January 2015 Revised 11 May 2015 Available online 27 May 2015 Keywords: Overhauser enhancement ESR Nitroxyl radical Carbamoyl-PROXYL Dynamic nuclear polarization Overhauser-enhanced magnetic resonance imaging

a b s t r a c t The dynamic nuclear polarization (DNP) studies were carried out for 15N labeled carbamoyl-PROXYL in pure water and pure water/glycerol mixtures of different viscosities (1.8 cP, 7 cP and 14 cP). The dependence of DNP parameters was demonstrated over a range of agent concentration, viscosities, RF power levels and ESR irradiation time. DNP spectra were also recorded for 2 mM concentration of 15N labeled carbamoyl-PROXYL in pure water and pure water/glycerol mixtures of different viscosities. The DNP factors were measured as a function of ESR irradiation time, which increases linearly up to 2 mM agent concentration in pure water and pure water/glycerol mixtures of different viscosities. The DNP factor started declining in the higher concentration region (3 mM), which is due to the ESR line width broadening. The water proton spin–lattice relaxation time was measured at very low Zeeman field (14.529 mT). The increased DNP factor (35%) was observed for solvent 2 (g = 1.8 cP) compared with solvent 1 (g = 1 cP). The increase in the DNP factor was brought about by the shortening of water proton spin–lattice relaxation time of solvent 2. The decreased DNP factors (30% and 53%) were observed for solvent 3 (g = 7 cP) and solvent 4 (g = 14 cP) compared with solvent 2, which is mainly due to the low value of coupling parameter in high viscous liquid samples. The longitudinal relaxivity, leakage factor and coupling parameter were estimated. The coupling parameter values reveal that the dipolar interaction as the major mechanism. The longitudinal relaxivity increases with the increasing viscosity of pure water/glycerol mixtures. The leakage factor showed an asymptotic increase with the increasing agent concentration. It is envisaged that the results reported here may provide guidelines for the design of new viscosity prone nitroxyl radicals, suited to the biological applications of DNP. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Nitroxyl radicals are less toxic and stable organic free radicals, which belong to the six-membered or five membered ring

Abbreviations: B0, Zeeman field; B0ESR, ESR Zeeman field; B0NMR, NMR Zeeman field; carbamoyl-PROXYL, 3-carbamoyl-2, 2, 5, 5-tetramethyl-pyrrolidine-1-oxyl; DNP, dynamic nuclear polarization; ESR, electron spin resonance; FID, free induction decay; FWHM, full-width half-maximum; GE, gradient echo; GP, gradient phase; GS, gradient slice; I, nuclear spin quantum number; MRI, magnetic resonance imaging; NMR, nuclear magnetic resonance; OMRI, Overhauser-enhanced magnetic resonance imaging; RF, radio frequency; S, electron spin quantum number; T1, water proton spin-lattice relaxation time; TE, echo time; TESR, ESR irradiation time; TR, Repetition time. ⇑ Corresponding author. E-mail address: [email protected] (A. Milton Franklin Benial). http://dx.doi.org/10.1016/j.jmr.2015.05.009 1090-7807/Ó 2015 Elsevier Inc. All rights reserved.

structure. These radicals have been widely used as a spin probe for invivo/invitro electron spin resonance (ESR)/Overhauserenhanced magnetic resonance (OMR) imaging techniques to analyze the interactions, reduction and oxidation status of free radicals and tissue oxygenation in living animals [1–9]. ESR imaging provides the spatial distribution of the nitroxyl radical where as anatomic information is not available. Overhauser-enhanced magnetic resonance imaging (OMRI) is a double resonance technique that uses the presence of paramagnetic agents to enhance the signal intensity from nuclear spins by means of a process known as dynamic nuclear polarization (DNP) or Overhauser effect [10,11]. In this phenomenon, the relatively stronger magnetic moment of the electron is used to enhance the polarization of nuclear spins, thereby enhancing their signal. The unique advantage of this technique is high spatial resolution of the image and short acquisition

M. Kumara Dhas et al. / Journal of Magnetic Resonance 257 (2015) 32–38

time. The significant contrast-to-noise ratio obtained by this technique makes OMRI advantageous in obtaining physiological information. OMRI is a promising technique for imaging the distribution and dynamics of free radicals [12–15], which is also used to demonstrate the reduction and oxidation process of 14N and 15N labeled nitroxyl radicals simultaneously [16]. Recently, Utsumi et al. used the membrane-impermeable spin probe, carboxy PROXYL and visualized the whole body kinetic image of mice with better spatiotemporal resolution, which provided both anatomic and physiological information of organ function for preclinical and clinical studies of oxidative diseases [17]. The nitroxyl spin probe, 3-carbamoyl-PROXYL has a partition coefficient that is intermediate between carboxy-PROXYL and methoxycarbonyl-PROXYL, which is expected to have intermediate level membrane permeable activity, and thus may penetrate cerebrovascular cell membranes with low efficiency. The nitroxyl spin probe 3-carbamoyl-PROXYL is one of the best spin probe for measuring nitroxide dynamics and redox imaging, based on magnetic resonance imaging of organic contrast agents in mice [18]. The carbamoyl-PROXYL, a redox sensitive water soluble spin probe was used to observe the small disruptions in brain vascular permeability by using magnetic resonance imaging technique [19]. The in vivo ESR/NMR co-imaging of the nitroxyl spin probe 3-carbamoyl-PROXYL in living mice enabled in vivo organ specific mapping of free radical metabolism and redox stress and the alternations occur in the parthogenesis of disease [20]. The isotopic substitution of the nitrogen atom from the naturally abundant 14 N to 15N provides enhanced detection sensitivity by decreasing the spectral multiplicity, the saturation of one ESR line in 15N nitroxyl radical induces a 17% increase in the proton polarization [21]. Hence, the 15N labeled carbamoyl-PROXYL nitroxyl radical is chosen as a spin probe for the present study. The DNP studies have provoked considerable interest in the field of biomedical imaging. The dynamic nuclear polarization properties of nitroxyl radicals in biological fluids was reported [22]. The dependence of image intensity on proton mobility was demonstrated for samples of the nitroxyl radical in water/glycerol mixtures of different viscosities [23,24]. Recently, the ESR studies on nitroxyl radicals in high viscous liquids have been reported, which reveals that the nitroxyl radicals having narrowest linewidth is suitable for in vivo/in vitro studies using ESRI/OMRI techniques [25]. In this work, OMRI has been used to produce NMR images reflecting changes in the viscosity of aqueous free radical solutions, with a view to studying molecular dynamics in biological samples such as blood, plasma, serum, plasma membrane and possibly in vivo. The 15N nitroxyl radical provides enhanced detection sensitivity by decreasing the spectral multiplicity. Hence the 15N labeled carbamoyl-PROXYL is chosen for this study. The biological fluids viscosity is in the range of 1.8– 14 cP. In order to understand and probe the viscosity effect on the DNP properties of 15N labeled carbamoyl-PROXYL in various solvents with different viscosities. Here, the dynamic nuclear polarization studies of 15N labeled carbamoyl-PROXYL in pure water and water/glycerol mixtures with different viscosities is reported.

2. Theory Theoretical principles of OMRI are well documented [26–28]. Nevertheless, a brief outline of the principles relevant to the experiments reported herein is presented in this section. The enhancement, E of the NMR signal of the 1H nuclei (I = 1/2) of water molecules with couplings to an unpaired electron spin S = 1/2 of a dissolved free radical, is given by

E¼

hIZ i jc j ¼ 1  qfs e I0 cN

33

ð1Þ

Here hIzi denotes the expectation value of the dynamic nuclear polarization, I0 is its thermal equilibrium value, q is the coupling parameter, f is the leakage factor, s denotes the saturation parameter, and ce and cN are, respectively the electron and nuclear gyro magnetic ratios. The leakage factor f in Eq. (1) that accounts for the loss of polarization, which is sensitive to the motion and it also depends upon the concentration of the nitroxyl agents, as given by,

f ¼1

T1 kCT 10 ¼ ; T 10 1 þ kCT 10

ð2Þ

Here T1 denotes the water proton spin–lattice relaxation time of the nitroxyl agent solution. Intrinsic nuclear relaxation rate of water proton in the absence of nitroxyl agent is denoted by 1/T10. The concentration of the nitroxyl agent is given by C, and k denotes the relaxivity constant. As the concentration of the agents is increased the leakage factor approaches to unity, because with increasing C, kC  1=T 10 The saturation factor (s) is critical to the sensitivity of OMRI, which is given by the degree of saturation of the electron spin,

s¼

ðS0  hSZ iÞ ; S0

ð3Þ

where S0 is the equilibrium polarization of the electron spins and hSZi is the polarization upon irradiation of the ESR resonance. The complete saturation occurs when the spin populations of the energy levels are equal and in this limit s approaches its maximum value of unity. For on resonance irradiation at the center frequency of one of the hyperfine components of the nitroxyl agent, by an oscillating magnetic field of amplitude B1, the saturation factor is given by,

s¼

c2e B21 T 1e T 2e 1 þ c2e B21 T 1e T 2e

ð4Þ

Here T1e and T2e are the electron spin–lattice relaxation time and spin–spin relaxation time, respectively. For complete saturation of one of the ESR transitions, and the condition that 1=T 10  1=T 1 , the enhancement factor is given by,

1E¼

ðce =cN Þq ð2I þ 1Þ

ð5Þ

where I is the relevant nuclear spin quantum number, (1 for 14N and ½ for 15N). The Overhauser enhancement reaches maximum values of 110 and 165, respectively for 14N and 15N nitroxyl agents for pure dipolar, and 220 and 330 for scalar interactions. Experimentally, these maximum values are not realized due to many factors. In nitroxyl agents, there is additional hyperfine interaction between the hydrogen nuclei and the unpaired electron. Hence the three (for 14N) or two (for 15N) ESR lines are inhomogeneously broadened due to the presence of the unresolved hydrogen hyperfine. An inhomogeneously broadened ESR line with very closely spaced hyperfine lines has Voigt line shape function, the convolution of a Lorentzian line shape with a Gaussian intensity profile [29]. Hence Eq. (4) will no longer hold good, and irradiation of one of the nitrogen hyperfine lines will result only in partial saturation. Nevertheless, the enhancement factor, achieved by irradiating a single ESR line of the nitrogen hyperfine line of the nitroxyl agent can be approximately given [23,30] by combining Eqs. (1), (2) and (5),

    1 1 1 1 1 ð2I þ 1Þ 1 þ 1þ ¼ 1  E 658 kCT 10 aP q

ð6Þ

Here P is the applied ESR power level, which is proportional to B21, and a is a constant related to the conversion efficiency of the coil and the relaxation times of the electron spins. Therefore, a plot of

34

M. Kumara Dhas et al. / Journal of Magnetic Resonance 257 (2015) 32–38

reciprocal enhancement against reciprocal power, 1/P, should give a straight line with intercept 1/(1  Emax) where Emax is the maximum enhancement achieved under complete saturation of the ESR signal. The effect of concentration on the enhancement factor can be better visualized by rearranging Eq. (6) as

  s 1 1 1 1þ ¼ 1  E 658 kCT 10 q

ð7Þ

where s ¼ ½1=ð2I þ 1Þ aP=ð1 þ aPÞ: Therefore, a plot of s/(1  E) against 1/C should be linear and the slope must be consistent with the independently determined kT10. 3. Materials and methods The spin probes, carbamoyl-PROXYL was purchased from Aldrich Chemical Co, St. Louis, MO, USA. 15N-labeled nitroxyl probes were synthesized in our laboratory by using 15N-ammonium chloride (Cambridge Isotope Laboratories, Inc. MA, USA) as per literature [21,30]. All other chemicals used were of commercially available reagent grade quality. The DNP experiments were performed at 23 °C on a custom-built (Philips Research Laboratories, Hamburg, Germany), human whole-body (magnet bore: 79 cm diameter; 125 cm length), low field (14.53 mT) scanner operating in a field-cycled mode to avoid excess RF power deposition during the ESR cycle [31]. The BESR was set at 6.57 mT for 15N labeled nitroxyl radical, 0 NMR and the B0 was set at 14.53 mT. The ESR irradiation frequency used was 220.6 MHz. A saddle coil (13.5 cm diameter, 23.5 cm length) was used as the ESR resonator. The efficiency parameter of the ESR coil used was measured to be 5.2 lT/W1/2. The NMR resonator assembly consisting of a transmit saddle coil (25 cm diameter, 23 cm length) and a receive solenoidal coil (5 cm diameter, 6 cm length) was tuned to 617 kHz with a band width of 1.5 kHz. To remove oxygen from the solution, argon gas was passed through the samples for about 2 h. The samples were prepared using the phosphate buffer solution at pH 7.4. The phantoms employed in the DNP experiments were 2 cm diameter tubes filled with 10 ml of various concentration of 15N labeled carbamoyl-PROXYL in pure water (solvent 1; g = 1 cP), pure water and glycerol mixtures of 80% and 20% by volume (solvent 2; g = 1.8 cP), pure water and glycerol mixtures of 50% and 50% by volume (solvent 3; g = 7 cP), pure water and glycerol mixtures of 40% and 60% by volume (solvent 4; g = 14 cP). The phantom tubes were placed vertically in the receive solenoidal coil. The enhancement factors were measured as a function of ESR irradiation time (TESR) for all the samples. DNP spectra of 2 mM concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 were recorded by sweeping the magnetic field (BESR 0 ) from 5 to 10 mT, in steps of 0.1 mT, while away from the resonance and in steps of 0.01 mT, while near the resonance for better peak identification using ESR irradiation time (TESR), 800 ms and power, 53.8 W. The ESR line width measurements were carried out at 23 °C for various concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 using an X-band ESR spectrometer (JEOL Co. Ltd., Akishima, Tokyo, Japan). The ESR linewidth measurements were carried out with an accuracy of ±0.5 lT. Imaging experiments were performed by using standard spin warp gradient echo sequence for MRI, except that each phase encoding step was preceded by an ESR saturation pulse to elicit the Overhauser enhancement. Fig. 1 shows the pulse sequence started with the ramping of the B0 field to 6.569 mT for 15N labeled nitroxyl radical, followed by switching on the ESR irradiation. Then, the B0 was ramped up to 14.529 mT before the NMR pulse (617 kHz) and the associated field gradients were turned on. At the beginning or end of the cycle, a conventional (native) NMR

Fig. 1. Field-cycled OMRI pulse sequence starts with the ramping of the BESR field to 0 6.569 mT for 15N-labeled nitroxyl radical, followed by switching on the ESR irradiation, (220.6 MHz). The B0 is ramped up to 14.53 mT before the NMR pulse and the associated field gradients are turned on. The field settling time was set at 18 ms.

signal intensity (with ESR OFF) was measured for computing the enhancement factors. A Hewlett–Packard PC (operating system, LINUX 5.2) was used for data acquisition. The images were reconstructed from the echoes by using standard software, and were stored in DICOM format (Digital Imaging and Communications in Medicine). MATLAB codes were used for the computation of DNP factors, curve fitting and water proton spin–lattice relaxation time. The reproducibility of the data was confirmed with several experiments and fitted parameters and enhancement factors show the good correlation (R2 > 0.999). Typical scan conditions were as follows, repetition time (TR)/echo time (TE): 2000 ms/25 ms; ESR irradiation time (TESR): 50 –800 ms, in steps of 50 or 100 ms; ESR power, 53.8 W; No. of averages, 10; phase encoding steps, 64 and slice thickness, 20 mm. The image field of view (48 mm) was represented by a 64  64 matrix, with a pixel size of 0.63 mm  0.63 mm. 4. Results and discussion 4.1. DNP spectra The DNP spectra of 2 mM concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 were measured by field-cycled DNP spectroscopy, by irradiating at a fixed ESR frequency (220.6 MHz) while sweeping the magnetic field to cover each resonance. When an ESR resonance condition is attained, the Overhauser effect causes an enhancement of the NMR signal, and its amplitude is altered. Therefore, a plot of NMR signal amplitude versus BESR shows the positions of the ESR resonances. 0 The relative amplitudes of the peaks provide information on the ESR signal intensities, modulated by the electron–proton coupling. The DNP spectra were collected with a field sweep of 5 mT for 15N, centered on 7.5 mT using the field-cycled DNP pulse sequence, which is shown in Fig. 2. The DNP spectra show two hyperfine lines for 15N-labeled carbamoyl-PROXYL. The hyperfine coupling of 15N is expected to be about 1.4 times that of 14N based on the nuclear gyro-magnetic ratio, the experimental hyperfine values from the DNP spectra agree well with the previous study [21]. The peak position and FWHM of DNP spectra of 2 mM concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 were calculated and listed in Table 1. In OMRI experiments, a saturating ESR pulse of duration TESR is applied preceding the NMR pulse. After switching on the ESR irradiation the nuclear polarization builds up at a rate governed by the water proton spin– lattice relaxation time, T1. To achieve the maximum possible Overhauser enhancement, it is necessary to irradiate the ESR transition of the nitroxyl agent for at least 3T1. But at high power, long TESR is an undesirable option for in vivo studies. Short TESR not only minimizes the heating effects but also reduces unwanted RF interference and hence limits the

M. Kumara Dhas et al. / Journal of Magnetic Resonance 257 (2015) 32–38

35

line width broadening is due to the dipolar and spin-exchange interactions, which in turn will decrease the saturation parameter and enhancement of NMR signal. Another factor that affects the build up and decay of the polarization of water protons is the NMR T1. Following the ESR irradiation, it is necessary to ramp up the magnetic field from the evolution field (6.569 mT) to the detection field (14.529 mT) in a time that is short compared to the sample’s T1. Hence it would be of interest to analyze how T1 is affected by the free radical concentration. Here, a simple method for proton T1 measurement, using OMRI has been recently reported [32]. This method is based on fitting the enhancement factor increases exponentially as a function of TESR using the following equation [30,32] Fig. 2. FC-DNP spectra of 2 mM concentration of 15N labeled carbamoyl-PROXYL in solvent 1 (s), solvent 2 (D), solvent 3 (h) and solvent 4 (). Pulse sequence parameters: TR/TESR/TE, 2000 ms/800 ms/25 ms; No. of averages, 2; other scan parameters are as given in the experimental section.

instrumental noise. Therefore, in the selection of TESR an optimal compromise needs to be made between the signal enhancement and the RF heating. In order to understand DNP phenomenon in liquid medium with different viscosities, the enhancement factors were measured for a moderate power of 53.8 W and ESR irradiation time (TESR) of 800 ms. The increased DNP factor (35%) was observed for solvent 2 compared with solvent 1. The increase in the DNP factor was brought about by the shortening of water proton spin–lattice relaxation time of solvent 2. The decreased DNP factors (30% and 53%) were observed for solvent 3 and solvent 4 compared with solvent 2, which is mainly due to low value of coupling parameter in high viscous liquids. 4.2. DNP factor The enhancement factors were measured as a function of ESR irradiation time for a moderate power of 53.8 W for various concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 by irradiating at the low field ESR line. The DNP parameters of 2 mM concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 were calculated and listed in Table 2. The enhancement factors were negative as usually reported in the DNP literature that takes account of the 180° phase shift of the FID when DNP occurs. Fig. 3 illustrates the effect of TESR on the DNP enhancement of 15N labeled carbamoyl-PROXYL of various concentrations in solvent 1, solvent 2, solvent 3 and solvent 4. The enhancement factor and ESR irradiation time fit with linear equation for lower agent concentration up to 2 mM agent concentration in pure water and pure water/glycerol mixtures of different viscosities. Further, the DNP factor started declining in the higher concentration region (3 mM), which is mainly due to the ESR line width broadening. Therefore, the agent concentration is optimized as 2 Mm for invitro/invivo studies. At higher concentration, the ESR

Table 1 Peak position and FWHM of DNP spectra of 2 mM solvent 1, solvent 2, solvent 3 and solvent 4. Samples

Solvent 1 Solvent 2 Solvent 3 Solvent 4

15

N labeled carbamoyl-PROXYL in

Peak position (mT)

FWHM (lT)

Low frequency

High frequency

Low frequency

High frequency

6.565 6.565 6.570 6.570

8.860 8.860 8.860 8.840

150 160 200 210

160 170 210 220

  T  ESR E  1 ¼ ðEinf  1Þ  1  exp T 1

ð8Þ

Here, E is the Overhauser enhancement factor, and Einf is the theoretical maximum enhancement factor corrected for the actual T1. By curve fitting the observed enhancement factors to Eq. (8), the water proton spin–lattice relaxation time for different concentrations of the free radical solutions was computed. The water proton spin–lattice relaxation time decreases with increasing agent concentration and viscosity. The DNP factor also depends on the water proton spin–lattice relaxation time (T1) and ESR irradiation time. The ESR irradiation time should be at least equivalent to 3T1 in order to achieve the maximum DNP effect. The NMR longitudinal relaxation time in solvent 2 (g = 1.8 cP) is shorter compared with the solvent 1, which enables the Overhauser enhancement 1.5 times compared with the solvent 1. In high viscous solvents, the ESR irradiation time TESR  3T1, but the decreased DNP factor was observed, which implies that the line width broadening plays major role in determing the DNP factor. The line width broadening is more in the high viscous solvents, which is due to the less mobile nature of the samples.

4.3. NMR relaxivity The relaxivity (k), and water proton spin–lattice relaxation time play an important role in the enhancement of NMR signal. The Eqs. (2) and (6) indicate the effect of free radical concentration on the enhancement factor. Eq. (2) reveals that the concentration dependence is via the leakage factor where the longitudinal NMR relaxivity (k) of the nitroxyl radical plays a significant role. The DNP factor is proportional to the free radical concentration as long as the concentration is low enough to cause any line broadening. Hence it would be of interest to estimate the relaxivity, k and water proton spin–lattice relaxation time in the absence of free radicals, T10. Fig. 4 shows the water proton longitudinal relaxation rate as a function of concentration for 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4. The relaxivity, k and water proton spin–lattice relaxation time in the absence of free radicals, T10 can be computed from the slope and intercept of the linear equation of the plot (Fig. 4), kT10 values are given in Table 2. The water proton spin–lattice relaxation time in the absence of free radicals, T10 for solvent 1 agrees well with the previous study [21]. The water proton spin–lattice relaxation time in the absence of free radicals, T10 for solvent 2, solvent 3 and solvent 4 decreases with the increasing viscosity of water/glycerol mixtures, which are given in Table 2. The relaxivity, k value increases with the increasing viscosity of the samples, which are listed in Table 2.

36

M. Kumara Dhas et al. / Journal of Magnetic Resonance 257 (2015) 32–38

Table 2 The DNP parameters of 2 mM concentration of

a b c

15

N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 respectively.

Samples

k (mM s)1

Emax  1c

DBppa (lT)

T1 (ms)

T10 (ms)

fb

q

kT10 from T1 measurement

Solvent Solvent Solvent Solvent

0.447 0.710 2.860 3.050

101.0 75.2 25.9 20.1

108.8 111.0 119.1 124.0

771 399 133 101

2237 1408 350 328

0.66 0.71 0.62 0.69

0.465 0.329 0.129 0.100

0.999 0.999 1 1

1 2 3 4

X-band, peak to peak line width for 2 mM concentration. f, leakage factor for 2 mM concentration. Emax is the extrapolated enhancement factor at complete saturation measured for 2 mM concentration.

Fig. 3. Illustration of the effect of TESR on the DNP enhancement of 15N labeled carbamoyl-PROXYL of various concentrations in solvent 1 (A), solvent 2 (B), solvent 3 (C) and solvent 4 (D). The solid Lines correspond to curve fit to the exponential decay equation. Scan parameters are as given in the experimental section.

4.4. ESR linewidth

4.5. Leakage factor

The ESR line width measurements were carried out for various concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4. Fig. 5 shows the ESR line width measured as a function of concentration for 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4. The ESR line width increases with the increasing agent concentration and viscosity of pure water/glycerol mixtures. In these mixtures, the free radical will interact with the water protons and glycerol OH protons. Therefore, this interaction also leads to the line broadening mechanism in the high viscous solvents. Fig. 5 shows the increase in ESR line width broadening above 2 mM concentration of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4. The ESR line width broadening leads to the different saturation parameter. The change in the enhancement factor was brought about by the saturation parameter. At higher concentration (3 mM), it is likely that dipolar interaction and spin-exchange interaction can broaden the ESR resonance, resulting in a decrease in saturation of the electron spin system for a given applied power, compared to that at lower concentrations, which in turn will reduce the OMRI signal enhancement. Therefore, the optimized agent concentration is 2 mM for in vivo/in vitro ESRI/OMRI experiments.

The leakage factor, f measures the fraction of the total relaxation rate due to nuclear-electron interactions. For weak interactions, f approaches zero and when the interactions are dominant, f = 1. The leakage factor (f) can be estimated from the water proton spin–lattice relaxation time in the absence of free radicals, T10 and the longitudinal relaxivity, k. Table 3 shows the values of leakage factor for various concentrations of 15N labeled carbamoylPROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 respectively. The leakage factor increases asymptotically with the increasing agent concentration of nitroxyl radicals. However, it is observed that above 2 mM the increase in f is not very significant. Above 2 mM, the line broadening will also reduce any enhancement achieved from the increased value of leakage factor. Hence the leakage factor values reveal that the optimized agent concentration is 2 mM for in vivo/in vitro ESRI/OMRI studies. The change in leakage factor (f) is not significant in the high viscous liquid samples. 4.6. Maximum enhancement factor (Emax) The enhancement factors were measured 2 mM of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3, and solvent 4, by varying the ESR irradiation power from 53.8 to 152.8 W.

M. Kumara Dhas et al. / Journal of Magnetic Resonance 257 (2015) 32–38

Fig. 4. The water proton longitudinal relaxation rate as a function of concentration for 15N labeled carbamoyl-PROXYL in solvent 1 (s), solvent 2 (D), solvent 3 (h) and solvent 4 ().

37

These results, plotted in Fig. 6 as reciprocal enhancement 1/(1  E), against reciprocal power, 1/P, give a linear relationship according to the Eq. (8). The intercept of a plot of 1/(1  E) against 1/P can be used to estimate Emax, the enhancement factor at complete ESR saturation (s = 1), which characterizes the overall interaction of the free radicals with the solvent. Therefore, from Fig. 6, on extrapolation to 1/P = 0, the Emax values were computed and listed in Table 2. For this investigation, a huge power level of up to 152.8 W was used to make the extrapolation more accurate. The maximum enhancement factor (Emax) decreases with the increasing viscosity of the samples. The decreased enhancement factors (30% and 53%) were observed for various concentrations of 15N labeled carbamoyl-PROXYL solvent 3 and solvent 4 compared with solvent 2, which is due to the ESR line width broadening and low value of saturation parameter. These results indicate that in the design of suitable viscosity prone spin probe for OMRI, importance must be given to factors that can enhance the saturation parameter (s) significantly.

4.7. Coupling parameter

Fig. 5. X-band ESR line width measured as a function of concentration for 15N labeled carbamoyl-PROXYL in solvent 1 (s), solvent 2 (D), solvent 3 (h) and solvent 4 ().

Table 3 The leakage factor for various concentrations of 15N labeled carbamoyl-PROXYL in solvent 1, solvent 2, solvent 3 and solvent 4 respectively. Samples

Agent concentration (mM)

Leakage factor (f)

Solvent 1

0.5 1 2 3 4

0.27 0.52 0.65 0.74 0.79

Solvent 2

0.5 1 2 3 4

0.36 0.57 0.71 0.79 0.83

Solvent 3

0.5 1 2 3 4

0.31 0.50 0.62 0.72 0.78

Solvent 4

0.5 1 2 3 4

0.29 0.50 0.69 0.75 0.80

The coupling parameter, q depends on the nature and time dependence of the nuclear-electron interactions. The coupling parameter takes the value 1, if the interactions are purely scalar, while q = 1/2, if the interactions are entirely dipolar. In bio-medical applications of DNP deals with low concentrations of small, rapidly tumbling free radical molecules in solution, the dipolar interactions are dominant, however the value of q has been found to be dependent on the actual type of free radical molecule [22]. The measurement of coupling parameter q (Eq. (1)) can throw light on the mechanism of coupling. The measurement of Emax enables the computation of q from Eq. (1). The q values were calculated and listed in Table 2. The observed coupling parameter, q has low value for 15N labeled carbamoyl-PROXYL in solvent 3 and solvent 4, which indicates that the week coupling between the electron spin with the nuclear spin. The low value of coupling parameter, q leads to the reduction in the enhancement factor. These coupling parameter values are in support of the dipolar interaction as the major mechanism of coupling between the electron spin with the nuclear spin, as observed in many other nitroxyl agents [22,23].

Fig. 6. Effect of ESR irradiation power on the enhancement for 2 mM of 15N labeled carbamoyl-PROXYL in solvent 1 (s), solvent 2 (D), solvent 3 (h) and solvent 4 (). The ESR power varied from 53.8 to 152.8 W for ESR irradiation time, 800 ms; other scan parameters are as given in the experimental section.

38

M. Kumara Dhas et al. / Journal of Magnetic Resonance 257 (2015) 32–38

5. Conclusions The dynamic nuclear polarization studies were carried out for N labeled carbamoyl-PROXYL in pure water and pure water/glycerol mixtures over a range of concentrations, viscosities, RF power level and ESR irradiation time. This study permits a clear understanding of the effect of molecular mobility upon Overhauser enhancement in high viscous liquid samples. The DNP factor started declining in the higher concentration region (3 mM), which is mainly due to the ESR line width broadening. Therefore, the agent concentration is optimized as 2 mM for in vivo/in vitro ESRI/OMRI experiments. The change in the enhancement factor was also brought about by the water proton spin–lattice relaxation time. The increased DNP factor (35%) was observed for solvent 2 compared with solvent 1. The observed coupling parameter, q has low value for 15N labeled carbamoyl-PROXYL in high viscous liquid samples (solvent 3 and solvent 4), which indicates that the weak coupling between the electron spin with the nuclear spin. The low value of coupling parameter, q leads to the reduction in the enhancement factor. The leakage factor, water proton spin–lattice relaxation time and longitudinal relaxivity were estimated. The viscosity was found to play an important role in the DNP mechanism. These results permit clear understanding of the factors determining enhancement and contrast in OMR images obtained by DNP. 15

Acknowledgments The authors A. Milton Franklin Benial, A. Jawahar and M. Kumara Dhas thank the college management for encouragement and permission to carry out this work. This work was supported by the Grant-in-Aid for JSPS Postdoctoral Fellowship for Foreign Researchers (ID No. P 04489). This work was also supported by the UGC Research Award scheme, New Delhi (F.No. 30-35/2011(SA-II)).

References [1] M.D. Lingwood, A.J. Sederman, M.D. Mantle, L.F. Gladden, Songi Han, NMR flow imaging, J. Magn. Reson. 216 (2012) 94–100. [2] M. Yamato, T. Shiba, T. Naganuma, K. Ichikawa, H. Utsumi, K. Yamada, Overhauser-enhanced magnetic resonance imaging characterization of mitochondria functional changes in the 6-hydroxydopamine rat model, Neurochem. Int. 59 (2011) 804–811. [3] O.V. Efimova, G.L. Caia, Z. Sun, S. Petryakov, E. Kesselring, A. Samouilov, Standard-based method for proton-electron double resonance imaging of oxygen, J. Magn. Reson. 212 (2011) 197–203. [4] V.V. Khramtsov, G.L. Caia, K. Shet, E. Kesselring, S. Petryakov, J.L. Zweier, A. Samouilov, Variables field proton-electron double resonance imaging: applications to pH mapping of aqueous samples, J. Magn. Reson. 202 (2010) 267–273. [5] C.L. Hawkins, M.J. Davies, Detection and characterisation of radicals in biological materials using EPR methodology, Biochim. Biophys. Acta 2014 (1840) 708–721. [6] P.M. Gonzalez, M.B. Aguiar, G. Malanga, S. Puntarula, Electron paramagnetic resonance (EPR) for the study of ascorbyl radical and lipid radicals in marine organisms, Compar. Biochem. Physiol. Part A 165 (2013) 439–447. [7] M. Elas, B.B. Williams, A. Parasca, C. Mailer, C.A. Pelizzari, M.A. Lewis, J.N. River, G.S. Karczmar, E.D. Barth, H.J. Halpern, Quantitative tumor oxymetric images from 4D electron paramagnetic resonance imaging (EPRI): methodology and comparison with blood oxygen level-dependent (BOLD) MRI, Magn. Reson. Med. 49 (2003) 682–691. [8] K. Matsumoto, M.C. Krishna, J.B. Mitchell, Novel pharmacokinetic measurement using electron paramagnetic resonance spectroscopy and simulation of in vivo decay of various nitroxyl spin probes in mouse blood, J. Pharmacol. Exp. Ther. 310 (2004) 1076–1083.

[9] M. Yamato, T. Egashira, H. Utsumi, Application of in vivo ESR spectroscopy to measurement of cerebrovascular ROS generation in stroke, Free Radic. Biol. Med. 35 (2003) 1619–1631. [10] M.C. Krishna, S. English, K. Yamada, J. Yoo, R. Murugesan, N. Devasahayam, J.A. Cook, K. Golman, J.H. Ardenkjaer-Larsen, S. Subramanian, J.B. Mitchell, Overhauser enhanced magnetic resonance imaging for tumor oximetry: coregistration of tumor anatomy and tissue oxygen concentration, Proc. Natl. Acad. Sci. USA 99 (2002) 2216–2221. [11] O.V. Efimova, Z. Sun, S. Petryakov, E. Kesselring, G.L. Caia, D. Johnson, J.L. Zweier, V.V. Khramtsov, A. Samouilov, Variable radio frequency proton– electron double-resonance imaging: application to pH mapping of aqueous samples, J. Magn. Reson. 209 (2011) 227–232. [12] R. Murugesan, S. English, K. Reijnders, K. Yamada, J.A. Cook, J.B. Mitchell, S. Subramanian, M.C. Krishna, Fluorine electron double resonance imaging for 19F MRI in low magnetic fields, Magn. Reson. Med. 48 (2002) 523–529. [13] H. Li, Y. Deng, G. He, P. Kuppusamy, D.J. Lurie, J.L. Zweier, Proton electron double resonance imaging of the in vivo distribution and clearance of a triaryl methyl radical in mice, Magn. Reson. Med. 48 (2002) 530–534. [14] N. Nestle, K. Shet, D.J. Lurie, Proton electron double resonance imaging of free radical distribution in environmental science applications— first results and perspectives, Magn. Reson. Imaging 23 (2005) 183–189. [15] T. Guiberteau, D. Grucker, EPR spectroscopy by dynamic nuclear polarization in low magnetic field, J. Magn. Reson., Ser. B 110 (1996) 47–54. [16] H. Utsumi, K. Yamada, K. Ichikawa, K. Sakai, Y. Kinoshita, S. Matsumoto, M. Nagai, Simultaneous molecular imaging of redox reactions monitored by Overhauser-enhanced MRI with 14N-and 15N-labeled nitroxyl radicals, Proc. Natl. Acad. Sci. USA 103 (2006) 1463–1468. [17] N. Kosem, T. Naganuma, K. Ichikawa, N.P. Morales, K. Yasukawa, F. Hyodo, K. Yamada, H. Utsumi, Whole body kinetic image of a redox probe in mice using Overhauser-enhanced MRI, Free Radical Biol. Med. 53 (2012) 328–336. [18] M. Yamato, T. Egashira, H. Utsumi, Application of invivo ESR spectroscopy to measurement of cerebrovascular ROS generation in stroke, Free Radical Biol. Med. 35 (2003) 1619–1631. [19] A. Tomizawa, I. Ishii, Z. Zhelev, I. Aoki, S. Shibata, M. Kitada, R. Bakalova, Carbamoyl-PROXYL-enhanced MRI detects very small distruptions in brain vascular permeability induced by dietary choesterol, Biochimica et Biophysics Acta 2011 (1810) 1309–1316. [20] R.M. Davis, S. Matsumoto, M. Bernardo, A. Sowers, K. Matsumoto, M.C. Krishna, J.B. Mitchell, Magnetic resonance imaging of organic contrast agents in mice. capturing the whole-body redox landscape, Free Radic. Biol. Med. 50 (2011) 459–468. [21] A.M.F. Benial, K. Ichikawa, R. Murugesan, K. Yamada, H. Utsumi, Dynamic nuclear polarization properties of nitroxyl radicals used in overhauserenhanced MRI for simultaneous molecular imaging, J. Magn. Reson. 182 (2006) 273. [22] D. Grucker, T. Guiberteau, B. Eclancher, J. Chambron, R. Chiarelli, A. Rassat, G. Subra, B. Gallez, Dynamic nuclear-polarization with nitroxides dissolved in biological-fluids, J. Magn. Reson. 106 (1995) 101–109. [23] I. Nicholson, D.J. Lurie, F.J.L. Robb, The application of proton–electron doubleresonance imaging techniques to proton mobility studies, J. Magn. Reson., Ser. B 104 (1994) 250–255. [24] P.L. de Sousa, R.E. de Souza, M. Engelsberg, L.A. Colnago, Mobility and free radical concentration effects in proton–electron doubleresonance imaging, J. Magn. Reson. 135 (1998) 118–125. [25] A.M.F. Benial, M.K. Dhas, A. Jawahar, Rotational correlation time studies on nitroxyl radicals using 300 MHz ESR spectrometer in high viscous liquid, Appl. Magn. Reson. 40 (2011) 311–319. [26] M.D. Lingwood, T.A. Siaw, N. Sailasutu, B.D. Ross, P. Bhattacharya, S. Han, Continuous flow Overhauser dynamic nuclear polarisation of water in the fringe field of a clinical magnetic resonance imaging system for authentic image contrast, J. Magn. Reson. 205 (2010) 247–254. [27] M.D. Lingwood, S. Han, Dynamic nuclear polarisation of 13C in aqueous solutions under ambient conditions, J. Magn. Reson. 201 (2009) 137–145. [28] S. Garcia, J.H. Walton, B. Armstrong, S. Han, M.J. McCarthy, L-band Overhauser dynamic nuclear polarisation, J. Magn. Reson. 203 (2010) 138–143. [29] J.H. Ardenkjaer-Larsen, I. Laursen, I. Leunbach, G. Ehnholm, L.G. Wistrand, J.S. Petersson, K. Golman, EPR and DNP properties of certain novel single electron contrast agents intended for oxymetric imaging, J. Magn. Reson. 133 (1998) 1– 12. [30] D.J. Lurie, J.M.S. Hutchison, L.H. Bell, I. Nicholson, D.M. Bussell, J.R. Mallard, Field-cycled proton electron double-resonance imaging of free-radicals in large aqueous samples, J. Magn. Reson. 84 (1989) 431–437. [31] G. Sosnovsky, Z.W. Cai, A study of the Favorskii rearrangement with 3-bromo4-oxo-2,2,6,6-tetramethylpiperidin-1-oxyl, J. Org. Chem. 60 (1995) 3414– 3418. [32] S. Matsumoto, H. Utsumi, T. Aravalluvan, K. Matsumoto, A. Matsumoto, N. Devasahayam, A.L. Sowers, J.B. Mitchell, S. Subramanian, M.C. Krishna, Influence of proton T1 on oxymetry using Overhauser enhanced magnetic resonance imaging, Magn. Reson. Med. 54 (2005) 213–217.