Magn Reson Mater Phy (2009) 22:327–331 DOI 10.1007/s10334-009-0187-x

SHORT COMMUNICATION

Detection of polyunsaturated omega-6 fatty acid in human malignant prostate tissue by 1D and 2D high-resolution magic angle spinning NMR spectroscopy Katarina Stenman · Jón B. Hauksson · Gerhard Gröbner · Pär Stattin · Anders Bergh · Katrine Riklund

Received: 2 April 2009 / Revised: 23 October 2009 / Accepted: 26 October 2009 / Published online: 17 November 2009 © ESMRMB 2009

Abstract Object Polyunsaturated omega-6 fatty acids (PUFAs) have been shown to promote prostate cancer. Here, we describe the use of HRMAS NMR spectroscopy to detect omega-6 PUFA species in prostate tissues. Materials and methods Samples originating from nonmalignant (n = 54) and malignant (n = 27) prostate tissues (from 27 prostatectomized men) were studied by 1D 1 H, 2D 1 H–1 H and 1 H–13 C HRMAS NMR spectroscopy followed by histopathological characterization. Results HRMAS NMR proved to be a powerful, nondestructive method to identify and characterize PUFAs. The omega-6 PUFA was found in 15% of examined human prostate tumors. Conclusion It is possible to detect PUFAs in prostate tissues using our NMR-based spectroscopic approach. K. Stenman (B) · K. Riklund Department of Radiation Sciences, Diagnostic Radiology, Umeå University and University Hospital of Northern Sweden, Umeå 901 87, Sweden e-mail: [email protected] J. B. Hauksson Radiation Physics, Umeå University and University Hospital of Northern Sweden, Umeå, Sweden G. Gröbner Department of Chemistry, Umeå University, Umeå, Sweden P. Stattin Department of Surgery and Perioperative Sciences, Urology and Andrology, Umeå University and University Hospital of Northern Sweden, Umeå, Sweden A. Bergh Department of Medical Biosciences Pathology, Umeå University and University Hospital of Northern Sweden, Umeå, Sweden

Keywords HRMAS NMR · Omega-6 & omega-3 fatty acids · PUFA · Prostate cancer · PCa Abbreviations GS Gleason score PCa Prostate cancer PUFA Polyunsaturated fatty acid HRMAS High-resolution magic angle spinning TOCSY Total correlation spectroscopy GE-HSQC Gradient enhanced heteronuclear single quantum correlation

Introduction Epidemiological and experimental evidence suggest that the incidence and mortality of prostate cancer (PCa) may be related to dietary factors [1]. The exact nature of the dietary factors involved is still not known, but various studies suggest that the amount and type of fat consumed, notably dietary contents of polyunsaturated fatty acids (PUFAs), may be related to prostate cancer progression [2–4]. There are two essentially different kinds of PUFAs, n-3 and n-6 types, both of which must be supplied through dietary intake [5]. The nature and strength of the relationships between PUFA intake, PUFA n-6 and n-3 levels, PCa incidence and its outcome are far from established [6], but most investigators believe that n-3 inhibits and n-6 stimulates PCa development [7]. Therefore, the aim of this study was to qualitatively investigate the presence of PUFAs in non-malignant and malignant prostate tissues, ex vivo, using HRMAS NMR. The non-destructive HRMAS NMR approach enables simultaneous analysis of hydrophilic compounds (e.g., citrate) as well as lipophilic compounds, such as PUFAs, and allows the analyzed specimens to be examined by light microscopy [8].

123

328

Magn Reson Mater Phy (2009) 22:327–331

Fig. 1 a NMR spectra obtained from a human PCa tissue selected for the intense resonance arising from diallylic protons at 2.8 ppm. b NMR spectrum from a randomly selected human non-malignant prostate tis-

sue displaying the characteristic double doublet arising from citrate at 2.75–2.55 ppm. The signals at 2.8 ppm are sub-baseline

However, a drawback of HRMAS NMR, compared to liquidstate NMR, is that there are more severe spectral overlaps between individual NMR resonances [9]. Therefore, 2D 1 H– 13 C correlation NMR experiments were performed in order to obtain the necessary spectral resolution for identification of the n-6 PUFA found in prostate tumors in our study. No n3 PUFA was detected spectrally in any of the prostate tissue samples analyzed.

and the following settings: 5 kHz sample spinning speed, 256 transients, 5 s recycle delay. Rotor-synchronized 2D TOCSY NMR experiments were performed with 41.8 ms mixing time at 2.25 kHz sampling and more sensitive (phase-sensitive) 2D HSQC 1 H–13 C NMR spectra were acquired at 5 kHz [10].The metabolite resonances and 2D HRMAS NMR cross peaks were assigned according to published chemical shift values [11,12]. After HRMAS NMR analysis, samples were fixed in formalin and embedded in paraffin, then 5-µm sections were stained with hematoxylin–eosin and subjected to histopathological analysis.

Materials and methods Tissue samples (one tumor and two non-malignant per individual) were obtained from 27 PCa patients (stage T1–T3, Gleason score 6 = 3 + 3 or 7 = 3 + 4) by a board-certified specialist in urogenital pathology (AB) within 20 min of surgical removal and immediately frozen in liquid N2 . Specimens were placed in 4 mm, tight MAS NMR rotors and upon addition of 10 µl D2 O analyzed by NMR at RT or stored at −20◦ C. The study was approved by the ethical review board (Umeå University, Sweden). NMR spectra were obtained using a 500 1 H MHz NMR instrument with a 4- mm HRMAS dual band (1 H and 13 C) probe (Bruker, Karlsruhe, Germany). 1D NMR spectra were acquired at RT using a rotor-synchronized Carr-Purcell-Meibom-Gill (CPMG) pulse sequence [90◦ -(τ -180◦ −τ )n -acquisition] with a T2 filter to suppress broad NMR resonances

123

Results The 1D HRMAS NMR spectra obtained from four of the 27 sampled tumor tissues (15%) displayed a very distinct 2.8 ppm resonance (Fig. 1a). Surprisingly, this NMR resonance was not detectable (i.e., levels were sub-baseline) in any of the 54 non-malignant prostate tissue samples (Fig. 1b). To assign the 2.8 ppm resonance to a specific metabolic component, additional TOCSY NMR experiments were carried out with the malignant human prostate samples. An illustrative example of the resulting spectra is shown in Fig. 2a. Protons arising from the molecular sequences CH=CH–CH2 – CH=CH (5.3 ppm), CH=CH–CH2 –CH=CH (2.8 ppm), CH=CH–CH2 –CH2 (2.1 ppm) and CH=CH–CH2 –CH2

Magn Reson Mater Phy (2009) 22:327–331

329

Fig. 2 a 2D HRMAS NMR TOCSY spectrum obtained from a human PCa tissue. The relevant regions for polyunsaturated fatty acid are shown, with the relevant cross peaks at 5.3, 2.8, 2.1 and 1.3 ppm. b 2D HRMAS NMR GE-HSQC spectrum obtained from human PCa

tissue: the pattern of encircled cross peaks indicates a PUFA of the n-6 type. No cross peak at 2.09/20.88 ppm, indicative of n-3 type PUFA, is visible

(1.3 ppm) are clearly visible and interconnected. This NMR pattern provides clear evidence that the 2.8 ppm diagonal peak originated from one or more PUFA species [12]. Further, 2D 1 H and 13 C correlated GE-HSQC NMR spectra indicated that the 2.8 ppm signal in the 1D NMR spectra arose solely from n-6 species [12], and there were no indications of the presence of n-3 species, the relevant NMR cross peaks showing the following 1 H/13 C chemical shifts: 2.79/26.06 ppm ( − CH2 − ), 2.07/26.17 ppm (-1 (H8,14)), 2.30/34.15 ppm (Fα(γ -chain)), 2.30/35.50 ppm (Fα(β-chain)), 1.61/25.50+25.30 ppm (Fβ), 1.30/32.37 ppm (Fω-3(-2)H15, 0.89/14.22 ppm (Fω(γ &β − chain)) (Fig. 2b). Cross peaks for n-3 would have had chemical shifts that were very close to those for n-6, but also one at 2.09/20.88 ppm (Fω-1(-1)) [12] which was absent in our NMR spectra (Figs. 2b and 3). The molecular structures of n-6 and n-3, showing the additional double bond of n-3 species that gives rise to the 2.09/20.88 ppm cross peak, can be seen in Fig. 3. The complementary GE-HSQC NMR spectra of non-malignant human prostate tissue did not indicate any presence of any PUFA species, as can be clearly seen in Fig. 3.

be the main culprit [3]. Notably, short-term dietary interventions have shown that low-fat diets with high intakes of plant products and daily supplements of fish oil (which has naturally high levels of n-3 fatty acids) can induce dramatic changes in gene expression patterns in human prostate tissue [13]. This indicates that prostate levels of multiple regulators of tissue growth and function may be affected by dietary factors. In particular, n-3 fatty acids such as α-linolenic acid (ALA, 18:3 n-3) and its derivatives eicosapentaenoic acid (EPA, 20:5 n-3) and docosahexaenoic acid (DHA, 22:6 n-3) seem to be positively linked to cancer prevention [14]. In marked contrast, n-6 species such as linoleic acid (LA, 18:2 n-6) and its derivatives γ -linolenic acid (GLA, 18:3 n-6), dihomo-GLA (DGLA, 20:3 n-6) and arachidonic acid (AA, 20:4 n-6) may be potent stimulants of prostate cancer development and aggressiveness [14]. Accordingly, our NMR study clearly shows that n-6 PUFAs accumulated in some malignant tissues, but were absent (or, more strictly, present at sub-detectable levels) in all of the non-malignant human prostate samples we examined. Presence of the n-6 PUFA also seemed to be related to higher Gleason scores, and possibly to higher tumor stages, since all four samples that generated NMR spectra displaying this PUFA were Gleason score 7 = 3 + 4 and showed extra-capsular growth (Table 1). Additional studies with a larger set of patients are needed to validate this hypothesis. The observation that n-6 PUFAs, as detected by HRMAS NMR, did not accumulate in the adjacent non-malignant

Discussion Several studies suggest that incidence of prostate cancer could be related to dietary factors, and that fat intake may

123

330

Magn Reson Mater Phy (2009) 22:327–331

Fig. 3 2D HRMAS NMR GE-HSQC spectrum obtained from non-malignant human prostate tissue lacking the cross peaks found in Fig. 2, which would indicate the presence of a PUFA of the n-6 type Table 1 Clinical characteristics of human PCa tissue from patients with and without the n-6 PUFA detected by NMR No. of patients

Preoperative PSA (no. of pat.)

Gleason score* (no. of pat.)

Pathological stage (no. of pat.)

3+3

T1

3+4

T2

T3

n-6 PUFA absent

23

8.9 ± 5.7

14

9

3

3

17

n-6 PUFA present

4

13.6 ± 6.4

0

4

0

0

4

∗

Significant differences in Gleason score between groups with and without PUFA present, P = 0.04 according to Fishers exact test

tissue in the prostate cancer patients we examined suggests that it could be more related to tumor growth than to premalignant stages of cancer development. It should be noted that HRMAS NMR detects freely rotating compounds in the cytoplasm, therefore PUFAs that are components of membranes cannot be detected by our approach [15]. In addition, the n-6 PUFA that was detected has not been observed in earlier magnetic resonance spectroscopy (MRS) PCa studies, to our knowledge. In an in vivo MRSI breast cancer study that aimed to detect PUFAs, wide between-patient variations in PUFA profiles were observed, but the clinical MR scanner used in the cited study was unable to distinguish between n-3 and n-6 PUFA species [16]. In general, essential fatty acids (EFA) are vital components of membrane lipids, they can only be obtained through dietary intake, and it is not straightforward to accurately control individuals’ diets. Furthermore, displacements and compartmentalizations of fatty acids naturally occur in cell membrane [15]. Thus, the variations in detectable levels of omega-6 (between- and within-subjects) and other PUFAs

123

may be due to various other factors, in addition to the known dietary factors, including variations in mitochondrial/cell membrane breakdown during cell death and/or fast cell turnover in lipid-rich regions.

Conclusions Omega-6 PUFA was detected in a subgroup of prostate cancer samples, in which omega-3 species were apparently absent, by 2D 1 H–1 H and 1 H–13 C HRMAS correlation NMR spectroscopy. HRMAS analysis of tissue biopsies may therefore be used to investigate the possible role of PUFAs in relation to PCa development and aggressiveness. Acknowledgments Support by the Lion’s Cancer Research Foundation (grant number AMP 04-408), Umeå University, the Swedish Cancer Foundation and Swedish Research Council is gratefully acknowledged. Thanks are due to Birgitta Ekblom, Elisabeth Dahlberg, Pernilla Andersson and Ingmar Sethson for technical assistance.

Magn Reson Mater Phy (2009) 22:327–331

References 1. Schumacher MC, Laven B, Wolk A, Brendler CB, Ekman P (2007) Do omega-3 die-tary fatty acids lower prostate cancer risk? A review of the literature. Curr Urol 1:2–10 2. Grönberg H (2003) Prostate cancer epidemiology. Lancet 361:850–864 3. Leitzmann MF, Stampfer MJ, Michaud DS, Augustsson K, Colditz GC, Willett WC, Giovannucci EL (2004) Dietary intake of n-3 and n-6 fatty acids and the risk of prostate cancer. Am J Clin Nutr 80(1):204–216 4. Ritch CR, Wan RL, Stephens LB, Taxy JB, Huo D, Gong EM (2007) Dietary fatty acids correlate with prostate cancer biopsy grade and volume in Jamaican men. J Urol 177(1):97–101 5. Das UN (2006) Essential fatty acids: biochemistry, physiology and pathology. Biotechnol J 1(4):420–439 6. Christensen JH, Fabrin K, Borup K, Barber N, Poulsen J (2006) Prostate tissue and leukocyte levels of n-3 polyunsaturated fatty acids in men with benign prostate hyperplasia or prostate cancer. BJU Int 97(2):270–273 7. Kelavkar UP, Lin Y, Landsittel D, Chandran, Dhir R (2005) The yin and yang of 15-lipoxygenase-1 and delta-desaturases: Dietary omega-6 linolenic acid metabolic pathway in prostate carcinogenesis. Prostaglandins Other Lipid Mediat 79(1–2):182–183 8. Griffin JL, Shockcor JP (2004) Metabolic profiles of cancer cells. Nat Rev Cancer 4(7):551–561 9. Swanson MG, Zektzer AS, Tabatabai L, Simko J, Jarso S, Dhir Keshari KR, Schmitt L, Carroll PR, Shinohara K, Vigneron DB, DhirKurhanewicz J (2006) Quantitative analysis of prostate metabolites using 1 H HR-MAS spectroscopy. Magn Reson Med 55(6):1257–1264

331 10. Parella T, Sánchez-Ferrando F, Virgili A (1997) Improved sensitivity in gradient-ba-sed 1D and 2D multiplicity-edited HSQC experiments. J Magn Reson 126(2):274–277 11. Swanson MG, Vigneron DB, Tabatabai L, Males RG, Schmitt L, Carroll PR, James JK, Hurd RE, Kurhanewicz J (2003) Proton HR-MAS spectroscopy and quantitative pathologic analysis of MRI/3D-MRSI-targeted postsurgical prostate tissues. Magn Reson Med 50(5):944–954 12. Willker W, Leibfritz D (1998) Assignment of mono- and polyunsaturated fatty acids in lipids of tissues and body fluids. Magn Reson Chem 36:S79–S84 13. Ornish D, Magbanua MJM, Weidner G, Weinberg V, Kemp C, Green C, Mattie MD, Marlin R, Simko J, Shinohara K, Haqq CM, Carrol PR et al (2008) Changes in prostate gene expression in men undergoing an intensive nutrition and lifestyle intervention. Proc Natl Acad Sci USA 105(24):8369–8374 14. Kobayashi N, Barnard RJ, Henning SM, Elashoff D, Reddy ST, Cohen P et al (2006) Effect of altering dietary omega-6/omega-3 fatty acid ratios on prostate cancer membrane composition, cyclooxygenase-2, and prostaglandin E2. Clin Cancer Res 12(15):4662– 4670 15. Griffin JL, Lehtimäki KK, Valonen PK, Gröhn OHJ, Kettunen MI, Ylä-Herttuala S (2003) Assignment of 1H nuclear magnetic resonance visible polyunsaturated fatty acids in BT4C gliomas undergoing ganciclovir-thymidine kinase gene therapy-induced programmed cell death. Cancer Res 63(12):3195–3201 16. He Q, Shkarin P, Hooley RJ, Lannin DR, Weinreb JC, Bossuyt VIJ (2007) In vivo MR spectroscopic imaging of polyunsaturated fatty acids (PUFA) in healthy and cancerous breast tissues by selective multiple-quantum coherence transfer (Sel-MQC): a preliminary study. Magn Reson Med 58(6):1079–1085

123