Introduction

Prostate Cancer (PCa) is the second leading cause of cancer death among men. Currently, an elevation of blood prostate specific antigen (PSA) level often results in surgical biopsies to determine if PCa is present. In the presence of PCa, biopsy cores are used to determine Gleason Scores (GS) through histopathology to assess PCa severity. However, the GS thus obtained from a random prostate biopsy often cannot accurately reflect PCa aggressiveness. Additional biomarkers are needed to assist precision medicine for individual PCa patients. The ongoing studies in our laboratory attempt to realize the potential utility of metabolite changes in the prostate due to PCa for more accurate diagnosis and patient prognosis. We have focused on quantifying prostate metabolites in intact tissue using high-resolution magic angle spinning magnetic resonance spectroscopy (HRMAS MRS) and performing metabolomic analyses to correlate tissue metabolic and metabolomic changes arising from PCa. Here we extend this work to evaluate metabolite diffusion in prostate and searched their correlations with PCa. Diffusion MRS can resolve different metabolites in a biological sample based on their sizes and shapes. Diffusion may be altered in different cellular compartments and this can be resolved using multiexponential analysis. Changes in metabolite diffusion may provide an additional dimension of metabolite profiling in PCa. We developed rotor-synchronized HRMAS Diffusion Ordered Spectroscopy (DOSY) for analysis of prostate tissue metabolite diffusions with radio-frequency pulses under field gradients.

Methods

Rotor-synchronized DOSY experiments were measured on a Bruker AVANCE III HD MR spectrometer at 4°C and under the HRMAS condition of 3600 Hz. Samples analyzed were: 10 mM solution of a prostate specific metabolite mixture, spermine and citrate; 20 mM agarose gel of spermine and citrate; intact prostate tissues from five PCa patients. DOSY measures metabolite signal intensities (I) as a function exponentially dependent on the strength of the pulse gradient (G): I ~ exp (c*G2), or ln(I) ~ c*G2, where c is an experimental constant that depends on the gyromagnetic ratio, the diffusion delays, as well as the diffusion coefficient (D). Therefore, by measuring metabolite intensities under different strengths of pulse gradients, D values of metabolites can be determined from the slopes of ln(I) ~ G2.

Results

Fig 1A presents spermine diffusion characteristics measured in the metabolite mixture solution, while Fig 1B shows the diffusion behavior of spermine in a prostate tissue sample. The variation of spermine diffusion coefficients can be estimated by using different experimental regions in the diffusion curve (the beginning, mid, and ending 20 spectra). In solution, the same regional analyses of the ln(I) vs G2 curve give the the same diffusion coefficient. However in tissue there is a clear deviation from linearity suggesting the existence of tissue spermine in multi-compartment cellular environments. Similar diffusion differences can also be seen between spermine in gel, which closely followed that in solution, and in tissue. In addition to spermine (Spm), the multi-compartment phenomenon could also be observed to various degrees for other tissue metabolites including citrate (Cit), creatine (Cre), lactate (Lac) and alanine (Ala), as shown in Fig 2. Metabolites in different sample diffuse differently likely due to their different micro-environments. Quantitative evaluations of diffusion coefficients measured for different metabolites in different tissues show strong intra-samples correlations among different metabolites, as shown in Fig 3 where the fast diffusion components as represented by the beginning 20 measured points (blue point in Fig 1, or values in Fig 2A) present significant linear correlations between citrate and spermine, and citrate and creatine. However, the diffusion rates are different for different metabolites with Cit:Cre=1:1, but Cit:Spm=4:3.

Discussion

Here, we present our initial diffusion (DOSY) studies of cellular metabolites in intact human prostate tissues and compared them with metabolite mixture solution and gel standard samples measured under the same experimental conditions. While our standard samples confirmed the ability of DOSY in quantifying unified metabolite diffusion coefficients in homogenous chemical and physical environments, the variations in diffusion coefficients seen for each metabolite in each tissue samples indicated the heterogenous environments within which they reside. These changes in diffusion based on compartmentalization present the potential for consideration as PCa markers. Quantification of other cellular metabolites; evaluation of their correlations with prostate pathologies; and modeling compartmentalization analysis are currently underway in our laboratory.

Conclusion

Our study has demonstrated the ability of DOSY in quantifying cellular metabolite diffusion coefficients from intact tissue with the assistance of the HRMAS method. Investigations of metabolite diffusion characteristics present the potential for more comprehensive understanding of disease metabolomic markers according to their cellular micro-environments.

Acknowledgements

We gratefully acknowledge the support of the Massachusetts General Hospital Athinoula A. Martinos Center for Biomedical Imaging.

References

No reference found.