Water structure differences between healthy and cancerous tissues    Return to Research page

Kenneth S Marsh1§, Thomas E Wagner 2, Carlton F Hazlewood3

1 Woodstock Institute for Science in Service to Humanity (WIS2H or WISSH), 130 Cane

Creek Harbor Road, Seneca, SC 29672

2 Greenville Hospital System, 900 W. Faris Road, Greenville, SC 29605

3 Research Consultants International, P.O. Box 130282, The Woodlands, TX 77393-0282

§Corresponding author             

Email addresses:   KSM: drksmarsh@aol.com    TEM: twagner@ghs.org    CFH: Carltonh@swbell.net


Water in different environments exhibits different responses in magnetic resonance (MR), and these differences allow MR to be used  for the diagnosis of disease, including cancer.  Some confusion existed in interpretation of results concerning differences arising from water within the cells and water in the extracellular matrix (water around cells).  This study concluded that water within cancer cells exhibits changes in MR response compared with healthy cells.  It is hoped that these measured differences may provide options for adjunctive therapies to exploit in future treatment of cancer.



Magnetic Resonance Imaging (MRI) is used to diagnose disease by presenting differences between healthy and diseased tissues.  MRI measures response of water protons, and therefore is presenting differences in the water environment.  Measurable differences in water structure between healthy and cancer cells could potentially be exploited for improved treatment of cancer if such differences resulted in altered solubility, drug action or sensitivities to radiation.  This study was designed to differentiate Magnetic Resonance response between water in the extracellular matrix and intracellular water in cancer and healthy cells.

Summary of Results

Cell cultures were prepared and analyzed through a chemical NMR with the aide of a contrast agent used to enhance medical MRI used to diagnose cancer.  The experiment led to an evaluation of the role of contrast agents to differentiate and diagnose cancer cells through the response of water within those cells, and to the conclusion that water in cancer cells is measurably less ordered than corresponding water in healthy cells.


Magnetic Resonance Imaging (MRI) is useful as a diagnostic tool because healthy and diseased tissues exhibit different contrast relationships in the images.  This is particularly true of cancer cells in which malignant tissue yields a brighter image than healthy tissue, as first demonstrated by Damadian in 1971.   Since the resonating entity in the MRI is primarily water protons, the contrast variation results from changes in the water environment (and water density) between malignant and healthy tissue (1), with the relaxation times of the water protons of the malignant tissue being significantly longer than that of the healthy tissue. Soon afterward, Hazlewood et al. (2) found that  premalignant stages could be recognized relative to hyperplastic and malignant stages. Diagnostic potential was demonstrated by further studies (3-5).   These differences could be recognized at the level of the organ, the tissues, and the cells (6,7) and the cellular organelles (8, 9). Carrying this even further, the relaxation times of water protons can be recognized relative to some initial state.  For example, division rate (10), phase of the cell cycle (11), in normal postnatal development of skeletal muscle (12), disease states other than cancer (13), and, conditions associated with filamentous changes within the cytoplasm (14).

Even within tissues, fractions of water can be recognized by magnetic resonance (MR) technology (a terminology use to include both NMR as well as MRI).  For example, Hazlewood, et al., (15) in agreement with Belton et al., (16), demonstrated at least three different fractions of tissue water which exhibit substantial differences in Nuclear Magnetic Resonance (NMR) transverse relaxation times (T2).  They described hydration water (approximately 8% of the cell water ) with T2 approximately 5.4 msec, myoplasm or cytoplasmic water (approximately 82% of the cell water) with T2 approximately 44 ms,  and extracellular water (approximately 10% of the cell water) with T2 approximately 155 ms.  Contrast these values against free or water in dilute electrolyte solutions with T2 values approximately 1.6 seconds. (Note: NMR and MRI are similar processes, with MRI modified to handle patients instead of small samples.)  These values are summarized in the following table.

Magnetic Resonance Response of Various Water Types

Water Description

% In cell

T2 Relaxation time

Water of Hydration


5.4  msecfillspace

Myoplasm or cytoplasm water


44.0  msecfillspace

Extracellular matrix water


155.0  msecfillspace

Free or dilute electrolyte solution water

1600.0  msecfillspace

                                                                                                                              (Hazlewood et al., 1974)

Considerable studies on the properties of water in various physical systems have suggested changes at interfaces with macromolecules (17 and 18).  Such changes could further modify the T2 values, resulting in a wide range of values which would be resolved as contrast differences in MR images (19).  

Response differences sufficient to yield contrasting images could potentially supply information about cancer cells that could be used to better understand the etiology of cancer.  It is conceivable that these very changes in the water environment that yield the imaging differential may be exploited in treatment.  With full recognition that the cause and effect have not been established, it is still possible that adjunctive agents or procedures, which alter the water environment, could potentially also alter the response to chemotherapeutic agents and/or radiation, hopefully to enhance their toxicity preferentially towards the malignant cell.   Protein conformation and water structural effects have already been related to relaxation times  (20 and 21).  We will refer to the influencing environment around the water protons collectively as water structure, and acknowledge that we can ascertain measurable differences, but have no definitive description of the morphology.

A fundamental concern is the point of origin of the different water structures.  Some controversies exist concerning the origins of the water response to MRI and the authors had different opinions concerning the response of intracellular water.  We acknowledged that water confined to the extracellular space (tissue area outside the cells) would exhibit a “freer” water (as demonstrated with a longer t2 relaxation time) in cancer cells.  Oedema has been observed and would support this conclusion.  However, we disagreed on whether contrast differences would also be generated by intracellular water in malignant versus healthy cells.

An experiment was therefore planned to resolve the discrepancy as described in Methods, below.   The experimental plan was conceptually simple; implementation was considerably more complex.

Experimental Difficulties:


A number of difficulties precluded successful completion of the experimental design:

              1) Cell culture requires growth of cells, which works well for malignant cells, but is considerably more difficult for healthy cells

              2) Cultures were prepared at the Greenville Hospital System which is a one hour drive from the NMR facility (The cells were kept viable during transport by taping them to the chest of the driver to maintain a suitable incubation temperature.)

              3) The operator of the chemical NMR was proficient in chemical studies, but less familiar with biological systems

              4) Cells adequately centrifuged to form a suitable pellet for this study could not be introduced into the NMR tubes; cells which could be introduced (not compacted) contained sufficient residual water to mask signals

              5) Gadolinium in this design suppressed all water signals

              6) As an initial, volunteer study, we had limited samples and NMR access.

Experimental Results:


The initial determination exhibited a water response that occluded any differences between samples.  The introduction of 0.05 cc gadolinium salt solution rapidly suppressed all signals.  This initial determination, therefore, was unable to resolve the controversy.  A repeat was not feasible at the time.

Post mortem


Although the experiment itself was inconclusive, a critical evaluation of the design enabled successful resolution of the original dispute and justification for the experiment.



The strength and weakness of this study is based upon a multidisciplinary approach that applies physical chemistry to biological systems.  Although new insights may result, the experimental protocol itself requires a substantial research effort that was beyond the realm of an exploratory study.  Conclusions can be drawn, however, to support the original hypothesis that intracellular water in malignant cells differs from water in healthy cells.

MRI results from proton alignment with a strong magnetic field and interactions with radio frequency pulses perpendicular to the magnetic field.  The interacting protons are primarily water protons with the second most abundant source arising from lipids.  Water protons therefore supply the major source for imaging. 

Gadolinium salts (Gd), such as Gadoiamide (gadolinium DTPA) are typically employed as contrast agents in medical imaging. Since Gd works rapidly to suppress a water signal, it is reasonable to surmise that in cellular systems, introduction of Gd before imaging suppresses the signal from extracellular water.  Medical studies performed during the approval of Gd contrast agents support this notion.  Caramia, et al. (22) states that paramagnetic materials containing unpaired electrons (such as Gd salts) induce relaxation of water protons, which allows greater signal intensity differentials to be imaged between diseased and normal tissues.  Harpur et al. (23) state “the pharmacokinetic behavior of gadodiamide was consistent with its extracellular distribution.”  Tweedle, et al. (24) state that Gd chelates with extracellular structures and that the hydrogen nuclei of water molecules that interact with the Gd center would relax more quickly (i.e. exhibit short relaxation times). This short relaxation time would reduce the signal from the chelated water which, for the most part would not be seen, thereby improving the contrast in the MR image.  Rapid renal elimination tends to maximize tolerance of Gd agents.  Rapid elimination of Gd chelates supports their primary action in the extracellular matrix rather than within the cell.

The appearance of contrast in the MR images suggests that the scan must be obtained before the Gd permeates the cells.  Therefore, the Gd suppresses the intercellular water signal (between the cells) but does not suppress the signal from the intracellular water (within the cells).  The fact that the images show contrasts that are useful in diagnosis must therefore be a manifestation of differences in the absorption of water protons within different cells.  Therefore, the success of Gd as a contrast agent arises as a direct result of its suppression of intercellular water signal which allows differences between intracellular water of different cells to be visualized.  Since this imaging is useful for cancer diagnostics, the intracellular water of malignant and healthy tissue must be different.

The major fraction of tissue water remains within the cells.  Therefore MR Imaging is useful even without contrast agents.  Oedema may also result in contrast difference between images of diseased and healthy tissue directly from the extracellular matrix.  But the effectiveness of contrast agents suggests differences of the intracellular water which may provide information to which could help lead to both better understand and treatment of cancer.

We make no additional claim for causation, but suggest that this explanation justifies investment in additional studies on water structure in cells and on biological agents which influence or modify water structure.


Water in cancer cells exhibits different response to MRI than water in healthy cells.  The contrast image differences, which represent longer relaxation times for cancer versus healthy cells, suggest that the water in healthy cells is more ordered than that in cancer cells, or conversely, water in cancer cells is more “bulk-like” than the structured water in normal cells.


A block of cells were prepared from a common organ, one set normal and the second malignant.  These cells were grown in cell culture.  Upon reaching a sufficient quantity of cells for NMR determination (target was 3 x 107/ml), the extracellular matrix was to be enzymatically removed, the cells centrifuged into a pellet which would then be scanned in a NMR spectrophotometer designed for chemical studies (9).  A chemical NMR was chosen because of availability and higher frequency (300 MHz as opposed to 3 MHz found in medical units) which would illustrate more subtle differences between samples.  Since residual water would exist outside the cells, a gadolinium salt (gadolinium DPTA) was employed to suppress the bulk water peak resulting from the outside water.  We had hoped that gadolinium introduced just prior to NMR analysis would suppress the residual extracellular water signal, but permeate the cell walls sufficiently slowly to not initially suppress signals from intracellular water.

Authors' contributions

The research protocol was developed by KM and TW.  CH supplied the background information and vital information concerning MRI.  The preliminary article was written by KM, and improved through the suggestions of TW and CH.  All authors read and approved the final manuscript.


We thank Greenville Hospital System, Greenville, SC, for supplying the cell cultures of both healthy and malignant tissue, Mountainview Medical Imaging, Seneca, SC, for their donation of the Gadolinium DTPA contrast agent, and Dr. Alex Kitaygorodskiy, Department of Chemistry, Clemson University for his valuable help with, and access to the Brucker 300 MHz NMR.  We also thank Amersham Health for background information on Gadolinium contrast agents.


      1. Damadian R., Tumor Detection by Nuclear Magnetic Resonance, Science 1971, 171, 1151.

     2. Hazlewood, C.F., Chang, D.C., Medina, D., Cleveland, G., and Nichols, B.L.  Distinction Between the Preneoplastic and Neoplastic State of Murine Mammary Glands.  Communicated by Norman Hackerman, April 6, 1972.  Proc. Nat. Acad. Sci.  1972, 69:1478-1480.

     3. Hazlewood, C.F., Cleveland, G., and Medina, D.  Relationship Between Hydration and Proton Nuclear Magnetic Resonance Relaxation Times in Tissues of Tumor-Bearing Mice:  Implications for Cancer Detection.  J. Nat. Cancer Inst. 1974, 52:1849-1852.

     4. Udall, J.N., Alvarez, L.A., Nichols, B.L., and Hazlewood, C.F.  The Effects of Cholera Enterotoxin on Intestinal Tissue Water as Measured by Nuclear Magnetic Resonance (NMR) SpectroscopyPhysiol. Chem. and Physics  1975, 7:533-539.

     5. Udall, J.N., Alvarez, L.A., Chang, D.C., Soriano, H., Nichols, B.L., and Hazlewood, C.F.  The Effects of Cholera Enterotoxin on Intestinal Tissue Water as Measured by Nuclear Magnetic Resonance (NMR) Spectroscopy IIPhysiol. Chem. and Physics 1977, 9:13-20.

     6.  Beall, P.T., and Hazlewood, C.F.  NMR Relaxation Times of Water Protons in Human Colon Cancer Cell Lines and ClonesCancer Biochem. Biophys. 1982, 6:7-12.

      7. Beall, P.T., Bonnie B. Asch, D. Medina and C.F. Hazlewood; Distinction of Normal, Preneoplastic, and Neoplastic Mouse Mammary Cells and Tissues by Nuclear Magnetic Resonance Techniques.  In:  The Transformed Cell.  Edited by Ivan C. Cameron and Thomas B. Pool. Academic Press, 1981, pp. 293-325.

     8. Michael, L.H., Seitz, P., McMillin-Wood, J., Chang, D.C., Hazlewood, C.F., and Entman, M.L. Mitochondrial Water in Myocardial Ischemia: Investigation with Nuclear Magnetic ResonanceScience 1980, 208:1267-1269.

     9. Kellermayer, M., Rouse, D., Gyorkey, F., and Hazlewood, C.F.  Ionic Milieu and Volume Adjustments in Detergent Extracted Thymic NucleiPhysiol. Chem. and Physics and Medical NMR 1983, 15:345-354.  

    10. Beall PT, Cailleau RM, Hazlewood CF, The relaxation Times of Water Protons and Division Rate in Human Breast Cancer Cells: A Possible Relationship to Survival, J. Chem. & Physics 1976, 8, 281.

    11. Beall, P.T., Hazlewood, C.F., and Rao, P.N.  Nuclear Magnetic Resonance Patterns of Intracellular Water as a Function of HeLa Cell CycleScience 192:904-907, 1976.

    12. Hazlewood, C.F., Nichols, B.L., Chang, D.C., and Brown, B.  On the State of Water in Developing Muscle:  A Study of the Major Phase of Ordered Water in Skeletal Muscle and its Relationship to Sodium Concentration.  The Johns Hopkins Medical Journal 128:117-131, 1971.

    13. Misra, L.K., Smith, N.K.R., Chang, D.C., Sparks, R.L, Cameron, I.L., Beall, P.T., Harrist, R., Nichols, B.L., Fanguy, R.C., and Hazlewood, C.F.  Intracellular Concentration of Elements in Normal and Dystrophic Skeletal Muscles of the Chicken.  J. Cellular Physiology 103:193-200, 1980.

    14. Beall, P.T., Brinkley, B.R., Chang, D.C., and Hazlewood, C.F.  Microtubule Complexes Correlated with Growth Rate and Water Proton Relaxation Times in Human Breast Cancer Cells.  Cancer Research 42:4124-4130, 1982.

     15. Hazlewood CF, Chang, DC, Nichols BL, Woessner, DE, Nuclear Magnetic Resonance Transverse Relaxation Times of Water Protons in Skeletal Muscle, Biophysical Journal 1974, 14, 583.

      16. Belton PS, Jackson RR, Packer KJ,  Pulsed NMR studies of water in striated muscle. I. Transverse nuclear spin relaxation times and freezing effects, Biochim Biophys Acta. 1972, 286(1):16-25 ..

      17. Drost-Hansen W, On the struture of water near solid interfaces and the possible existence of long range orderInd. Eng. Chem. 1969, 61, 10-47.

      18. Drost-Hansen W, Structure and properties of water at biological interfaces, in Chemistry of the Cell Interface, Part B, Edited by Brown HD, Acdemic Press, Inc., New York 1971, 6:1-184.

      19. Kasturi SR, Hazlewood CF, Yamanashi WS, Dennis LW, The Nature and Origin of Chemical Shift for Intracellular Water Nuclei in Artemia Cysts, Biophs. J.  1987, 52, 249-256.

     20. Rorschach HE, Hazlewood CF, Protein Dynamics and the NMR Relaxation Time T1 of Water in Biological ScienceJournal Magnetic Resonance 1986, 70:79-88. 

      21.  Egan TF, Rorschach HE,  Hazlewood CF, Molecular Basis of Contrast in MRI,  In: Cell Function and Disease,  Edited by Cañedo LE. Todd LE, Packer L,  Jaz J,  Plenum Press, New York, 1988,  pp. 405-413 .

     22. Caramia F, Aronen HJ, Sorensen AG, Belliveau JW, Gonzalez RG, Rosen BR, Perfusion MR Imaging with Exogenous Contast Agents,  In Diffusion and Perfusion Magnetic Resonance Imaging Edited by Le Bihan D. Raven Press, New York 1995, 14:255-267.

     23. Harpur ES, Worah D,  Hals PA, Holtz E, Furuhama K, Nomura H,  Preclinical safety assessment and pharmacokinetics of gadodiamide injection, a new magnetic resonance imaging contrast agent, Invest Radiol., 1993 Mar;28 Suppl 1:S28-43. Review.

     24. Tweedle  MF.,  Wedeking P, Kumar K, 1995, Biodistribution of radiolabeled, formulated gadopentetate, gadoteridol, gadoterate, and gadodiamide in mice and rats, Invest Radiol. 1995 Jun; 30(6):372-80.