Jack S. Cohen

Early Detection of Response to Radiation Therapy in Patients With Brain Malignancies Using Conventional and High b-Value Diffusion-Weighted Magnetic Resonance Imaging

PURPOSEnTo study the feasibility of using diffusion-weighted magnetic resonance imaging (DWMRI), which is sensitive to the diffusion of water molecules in tissues, for detection of early tumor response to radiation therapy; and to evaluate the additional information obtained from high DWMRI, which is more sensitive to low-mobility water molecules (such as intracellular or bound water), in increasing the sensitivity to response.nnnPATIENTS AND METHODSnStandard MRI and DWMRI were acquired before and at regular intervals after initiating radiation therapy for 10 malignant brain lesions in eight patients.nnnRESULTSnOne week posttherapy, three of six responding lesions showed an increase in the conventional DWMRI parameters. Another three responding lesions showed no change. Four nonresponding lesions showed a decrease or no change. The early change in the diffusion parameters was enhanced by using high DWMRI. When high DWMRI was used, all responding lesions showed increase in the diffusion parameter and all nonresponding lesions showed no change or decrease. Response was determined by standard MRI 7 weeks posttherapy. The changes in the diffusion parameters measured 1 week after initiating treatment were correlated with later tumor response or no response (P <.006). This correlation was increased to P <.0006 when high DWMRI was used.nnnCONCLUSIONnThe significant correlation between changes in diffusion parameters 1 week after initiating treatment and later tumor response or no response suggests the feasibility of using DWMRI for early, noninvasive prediction of tumor response. The ability to predict response may enable early termination of treatment in nonresponding patients, prevent additional toxicity, and allow for early changes in treatment.

Interaction of oligonucleotides with cationic lipids: the relationship between electrostatics, hydration and state of aggregation

Lipoplexes, which are spontaneously formed complexes between oligonucleotide (ODN) and cationic lipid, can be used to deliver ODNs into cells, both in vitro and in vivo. The present study was aimed at characterizing the interactions associated with the formation of lipoplexes, specifically in terms of electrostatics, hydration and particle size. Large unilamellar vesicles (approximately 100 nm diameter), composed of either DOTAP, DOTAP/cholesterol (mole ratio 1:1) or DOTAP/DOPE (mole ratio 1:1) were employed as a model of cationic liposomes. Neutral vesicles ( approximately 100 nm diameter), composed of DOPC/DOPE (mole ratio 1:1), were employed as control liposomes. After ODN addition to vesicles, at different mole ratios, changes in pH and electrical surface potential at the lipid-water interface were analyzed by using the fluorophore heptadecyl-7-hydroxycoumarin. In separate mirror image experiments, liposomes were added at different mole ratios to fluorescein isothiocyanate-labeled ODNs, thus yielding data about changes in the pH near the ODN molecules induced by the complexation with the cationic lipid. Particle size distribution and turbidity fluctuations were analyzed by the use of photon correlation spectroscopy and static light-scattering, respectively. In additional fluorescent probe studies, TMADPH was used to quantify membrane defects while laurdan was used to measure the level of hydration at the water-lipid interface. The results indicate that mutual neutralization of cationic lipids by ODNs and vice versa is a spontaneous reaction and that this neutralization is the main driving force for lipoplex generation. When lipid neutralization is partial, induced membrane defects cause the lipoplexes to exhibit increased size instability.

Oxidative stress causes membrane phospholipid rearrangement and shedding from RBC membranes--an NMR study.

Nuclear Magnetic Resonance (NMR) spectroscopy was used to investigate the relationship between oxidative stress experienced by RBCs and their phospholipid content and shedding. Using 1H-NMR, we demonstrated a higher lactate/pyruvate ratio, an indicator of oxidative stress, in normal RBCs treated with oxidants (t-butylhydroxyperoxide and H2O2) as well as in beta-thalassemic RBCs. Using 31P-NMR, we found 30% more phosphatidylcholine (PC), and unexpectedly, 35% less phosphatidylserine (PS) in the thalassemic RBCs. PS was decreased by treatment with oxidants and increased by anti-oxidants (vitamin C and N-acetyl cysteine); PC showed the opposite behavior. Thalassemic RBCs incubated in phosphate buffered saline produced more PS in the supernatant than normal RBCs. Anti-oxidants reduced the PS in the supernatant while oxidants increased it. Plasma of thalassemic patients contained 2.6-fold and 1.8-fold more PS and PC, respectively, than normal plasma. These results indicate that the decreased PS in RBCs resulted from increased shedding. The nature of the shed PS was studied by purifying and analyzing membranous microparticles from the plasma and RBC supernatants. More PS was found in microparticles purified from thalassemic plasma and RBC supernatants (5.6- and 4.8-fold, respectively) than in their normal counterparts. However, the bulk (80-90%) of the shed PS was not associated with microparticles. The significance of PS shedding for RBC survival needs further clarification.

Oligonucleotide lipoplexes : the influence of oligonucleotide composition on complexation

Despite extensive investigations into oligonucleotide lipoplexes, virtually no work has addressed whether the physicochemical properties of these assemblies vary as a function of the constituent oligonucleotide (ODN) sequence and/or composition. The present study was aimed at answering this question. To this end, we complexed N-(1-(2,3-dioleoyloxy)propyl)-N,N,N-trimethylammonium chloride (DOTAP) liposomes, in dispersion, with either 18-mer phosphorothiote homo-oligonucleotides composed of either adenine, thymidine or cytosine; or one of three structurally related 18-mer phosphorothioate oligonucleotides (S-ODNs) (G3139, its reverse sequence and its two-base mismatch). After ODN addition to vesicles at different mole ratios, changes in pH and electrical surface potential at the lipid-water interface were analyzed by using the fluorophore heptadecyl-7-hydroxycoumarin while particle size distributions were analyzed by static-light scattering. The results indicate that each homo-oligonucleotide does indeed exhibit different complexation behavior. In particular, the maximal level of DOTAP neutralization by the polyadenine S-ODN is much lower than that for the two other homo-oligonucleotides and hence its lipoplex is much more positively charged. Much smaller electrostatic differences are also apparent between lipoplexes formed from each of the G3139-related ODNs. This paper identifies nucleotide base selection and sequence as a variable that can affect the physicochemical properties of oligonucleotide lipoplexes and hence probably their transfection competency.

Hormone sensitivity is reflected in the phospholipid profiles of breast cancer cell lines.

We have found that the profiles of total phospholipids in malignant breast cancer cell lines change going from hormone sensitive to highly hormone resistant cells lines. In particular, two phospholipid components that were absent or at very low levels in hormone sensitive MCF7 cells and moderately hormone sensitive cell lines (MIII, LCC2) were found in relatively high proportions in highly hormone resistant cell lines (MB435, MB231). These two components were shown to be the alkylacylphosphatidylcholine (AAPtdC) and the unsaturated analog plasmenylphosphatidylethanolamine (plasmenyl-PtdE). Another component phosphatidylethanolamine (PtdE) increased in correlation with the degree of hormone insensitivity. This was shown using 31P NMR spectroscopy of lipid extracts of the cells, and was confirmed using HPLC analysis, as well as other techniques. The significance of these results for the metabolic characteristics of these cell lines is related to the therapeutic responsiveness of breast cancer.

Specific lipoplex-mediated antisense against Bcl-2 in breast cancer cells: a comparison between different formulations

G3139 is an antisense oligonucleotide (ODN) that can down-regulate bcl-2, thus potentially acting as a potent anticancer drug. However, effective therapy requires efficient ODN delivery, which may be achieved by employing G3139 lipoplexes. Yet, lipofection is a complex, multifactorial process that is still poorly understood. In order to shed more light on this issue, we prepared 18 different G3139 lipoplex formulations and compared them in terms of their capability to transfect MCF-7 breast cancer cells. Each formulation was composed of a cationic lipid and sometimes a helper lipid. The cationic lipid was either DOTAP (N-(1-(2,3-dioleoyloxy)propyl)-N,N,N-trimethylammonium chloride), DC-CHOL (3β[N-(N′,N′-dimethylaminoethane)carbamoyl]-cholesterol), or CCS (ceramide carbomoyl spermine). The helper lipid was either DOPC, DOPE, or cholesterol. Each lipid combination existed in two different structural forms — either large unilamellar vesicles (∼100 nm LUV) or unsized heterolamellar vesicles (UHV). Cell proliferation assays were used to evaluate the cytotoxicity of G3139 lipoplexes, control cationic lipid assemblies, and free G3139. Western blots were used to confirm the specific activity of G3139 as an anti-bcl-2 antisense agent. We determined that treatment of MCF-7 cells with G3139:CCS lipoplexes (UHV-derived) produced a maximal 50-fold improvement in antisense efficacy compared to treatment with free G3139. The other G3139 lipoplexes were not superior to free G3139. Thus, successful lipofection requires precise optimization of lipoplex lipid composition, structure, and concentration.

Determination of intracellular pH and compartmentation using diffusion-weighted NMR spectroscopy with pH-sensitive indicators

The intracellular pH (pHi) of a series of cancer cell lines was determined using the pH‐sensitive indicators imidazole (Im) or histidine (His) and diffusion‐weighted (DW) proton NMR spectroscopy. The DW method allows the observation at high magnetic field gradient values of only the slow‐moving (intracellular) components, thus ensuring complete separation between intra‐ and extracellular components. Using the chemical shift difference (Δδ) between the imidazole ring C2‐H and C4(5)‐H peaks, we were able to measure the pHi independently of chemical shift standardization. With His, the cell lines gave pHi values of ∼6.5–7.0, whereas with Im, a second, more acidic compartment (pHi = 5.5–5.8) was also observed. An inverse correlation was also found between pHi and the intracellular lactate concentration. This method may be applicable to in vivo pH determinations. Magn Reson Med 51:900–903, 2004.

31P-Magnetic resonance spectra of ovarian cancer cells exposed to chemotherapy within a three-dimensional Matrigel construct

We aimed to determine the metabolic profile and effects of chemotherapy on ovarian cancer cell metabolism in a three-dimensional (3D) vs. a two-dimensional (2D) construct using 31P-magnetic resonance spectroscopy (MRS). Three ovarian cancer cell lines were embedded in a 3D perfused Matrigel construct or grown in a 2D monolayer. Metabolic differences between the three cell lines were determined using 31P-MRS both in the 3D and in the 2D constructs. Cells were incubated with three different cytotoxic drugs at LC50 for 44xa0h and evaluated for metabolic changes using 31P-MRS. While the 3D construct allowed MRS assessment of viable cells, the 2D monolayer permitted evaluation of non-viable cell extracts. In both cells embedded in Matrigel (CEM) and cells grown in monolayers (CGM) different cancer cell lines showed characteristic metabolic fingerprints, which differed significantly between CEM and CGM. In contrast to the cell monolayer, CEM allowed continuous monitoring of the changes in 31P-MRS spectra over time following exposure to chemotherapy, demonstrating a progressive decrease in specific phosphorylated metabolites. The metabolic response of CEM and CGM to various antimitotic agents was significantly different. We conclude that different ovarian cancer cell lines show characteristic 31P-MRS fingerprints and specific metabolic changes in response to cytotoxic drug treatment. The perfused 3D Matrigel construct is superior to the 2D tissue monolayer for 31P-MRS studies, because it simulates the inxa0vivo conditions more closely and facilitates MRS evaluation of viable cells as well as continuous monitoring of metabolic changes in response to chemotherapy over time.

Biomedical applications of diffusion-weighted NMR techniques

MRI is a well-established tool in clinical diagnosis, but it lacks the specificity that would allow clear distinctions between many disease conditions. That is mainly because it is a spatial tool that lacks functional information. There are several other NMR techniques that could be used to provide functional information and thus improve the specificity of MR diagnostics. One of these is the use of diffusion-weighted (DW) MRI, which provides information about the diffusion of water in tissue. Preliminary animal and clinical results indicate that changes of the diffusion properties in tumors might provide early prediction of response or non-reponse to therapy. Another useful adjunct to MRI is the application of NMR spectroscopy (or MRS), which provides spectroscopic information mainly on the metabolic components of the tissue. However, notwithstanding many years of research, proton MRS has not so far proved a useful adjunct to MRI for the physician. This is mainly because of (a) the presence of the huge solvent water peak and (b) the mixture of both intra- and extracellular metabolic components in the tissue. A further extension of MRS would be the use of DWMRS, which is the application of the DW technique to the spectroscopic components of tissue other than water. Using DWMRS we are able to observe only the intracellular metabolic components and their changes due to drug treatment. In this article we will explore the potential of these DW methods as applied to (a) perfused cells, (b) brain tumors in mice, and (c) human brain tumors.

Early non-invasive detection of brain tumor response to radiation using diffusion weighted MRI

patient ranged from 0.5cm to 2.6 cm, with a median of 1.6 cm. This difference was not related to patient’s AP and right-left dimensions. The lateral distances from the anterior midline are: 3.5 ,-0.4; 2.9 ,-0.4; 2.5 ,-0.4; 2.9 ,-2.7;, and 2.7 ,-0.7 cm, at the 1st, 2nd, 3rd, 4th, and 5th intercostal spaces respectively. Although the IM lateral positions varied from patient to patient, they were relatively consistent in respect to the lateral border of sternum, all located maximally less than 1 cm laterally at 2nd, 3rd, 4th, and 5th intercostal spaces, and 0.5 cm medially of the 1st intercostal space. Distances from the most lateral border of AXN and the most medial border of HH were -2.1 ,-1.3, 0.7 ,-0.5, and 2.9 ,-1.5 cm at levels of HH head, coracoid process and lateral end of clavicle respectively (negative value indicates overlapping). AXN frequently overlapped HH anteriorly when the arm was angled more than 90 degree, but did not when the arm is placed 90 degree or less. The gantry angle of the SCV port also effected HH sparing. The larger the angle, the less HH could be spared. DVHs of IMC, AXN, and HH at different conventional beam settings were also analyzed and showed significant variation of node coverage and HH sparing.