Elsevier

NeuroImage

Volume 89, 1 April 2014, Pages 271-279
NeuroImage

Assessment of rodent brain activity using combined [15O]H2O-PET and BOLD-fMRI

https://doi.org/10.1016/j.neuroimage.2013.11.044Get rights and content

Highlights

  • A novel [15O]H2O-PET imaging protocol reveals brain activation in rodents.

  • First comparison study of [15O]H2O-PET and BOLD-fMRI in small animals

  • A difference between these two brain mapping methods was observed.

  • Discrepancy in terms of location of peak activation centers and spatial extend

  • The discovered results point towards a complementary nature of both techniques.

Abstract

The study of brain activation in small animals is of high interest for neurological research. In this study, we proposed a protocol to monitor brain activation in rats following whisker stimulation using the short half-life PET tracer [15O]H2O as a marker for cerebral blood flow. This technique enables the study of baseline and activation conditions in fast succession within the same scanning session. Furthermore, we compared the results obtained from PET imaging with additional BOLD-fMRI data acquired in the same animals within the same anesthetic session in immediate succession. Although the maximum relative signal changes during brain activity observed with PET were substantially higher compared to the BOLD-fMRI results, statistical analyses showed that the number of activated voxels in PET was lower compared to the fMRI measurements. Furthermore, there was a difference in the activation centers in both the shape and location between PET and fMRI. The discrepancy in the number of activated voxels could be attributed to a lower overall contrast-to-noise ratio of the PET images compared to BOLD-fMRI, whereas the difference in the spatial location indicates a more fundamental process, involving the different physiological origins of the PET and BOLD-fMRI response. This study clearly demonstrates that [15O]H2O-PET activation studies may be performed in small laboratory animals, and shows the complementary nature of studying brain activation using [15O]H2O-PET and fMRI.

Introduction

The study of the functional processes of the brain is of high interest for basic research and for clinical diagnosis. Positron emission tomography (PET) (Frackowiak and Friston, 1994) using [15O]H2O (half-life time T1/2 = 122 s) as a marker for cerebral blood flow (CBF) to map brain activation has been widely employed to investigate brain function in humans (Fox and Mintun, 1989, Fox and Raichle, 1986, Hummel et al., 2009, Payoux et al., 2010) because it allows a rapid alteration between baseline and activation scans due to the short half-life of [15O]. The blood oxygen level dependent effect (BOLD), which has been used more recently in functional magnetic resonance imaging (fMRI) studies of brain activation, reflects the complex interplay between changes in CBF, cerebral blood volume (CBV), cerebral metabolic rate of oxygen consumption (CMRO2) and oxygen extraction fraction (OEF) (Buxton, 2010). Despite the widespread utilization of the BOLD effect to measure brain activation in humans (Brown et al., 2011, Iannetti and Wise, 2007, Luijten et al., 2011) and in animals (Just et al., 2010, Sanganahalli et al., 2008, Seehafer et al., 2010), its physiological bases are not entirely understood (Buxton, 2010, Logothetis et al., 2001, Shulman et al., 2007). Thus, it would be of the utmost interest to measure and compare brain activation using both [15O]H2O-PET and BOLD-fMRI to cross-validate these two markers of brain function and to deconvolute the complex nature of the BOLD signal.

Brain activation studies in rodents undergoing sensory stimulation and assessed using PET in combination with the tracer [15O]H2O have, to the best of our knowledge, not been previously described. There have been a few CBF measurements performed in rats using PET, and most examine global values (Ose et al., 2012, Watabe et al., 2013, Yee et al., 2005). Previous studies comparing PET and BOLD-fMRI have only been performed in humans and have produced contradictory results. Ramsey and colleagues compared [15O]H2O-PET with BOLD-fMRI in sequential measurements, where they identified a high correlation in terms of sensitivity between the two methods (Ramsey et al., 1996). Kinahan et al. found a significant mismatch in spatial location between the PET and fMRI activation centers in humans (Kinahan and Noll, 1999). In addition, Joliot et al. also found a mismatch between the PET and fMRI activation maps in humans (Joliot et al., 1999), whereas a subsequent experiment by Devlin et al. concluded similar, but not identical, PET and fMRI results (Devlin et al., 2000).

To shed additional light on to the relationship between PET and fMRI activation studies, we used a small animal model, which can be extensively examined under more controlled conditions and multiple repetitions. The aim of this study was to compare [15O]H2O-PET and BOLD-fMRI measurements in rats that were acquired in immediate succession during the same anesthetic session using a whisker stimulus. To achieve this goal, we established a noninvasive in vivo brain activation mapping protocol in small animals using [15O]H2O-PET.

Section snippets

Animal preparation

All of the animal experiments were approved by the local authorities (Regierungspraesidium Tuebingen). Eight male Lewis rats at an age of 17 ± 4 weeks and a weight 358 ± 16 g (Charles River Laboratories, Sulzfeld, Germany) were used in the combined PET and MR brain activation measurements. Each animal was initially anesthetized with a mixture of 1.5% isoflurane vaporized in air at a gas flow rate of 1.0 L/min. The animals were placed head first in a prone position on a multimodality imaging bed

Coregistration accuracy of the PET/MR atlas

The generated PET and MR atlases (Supplementary Fig. 3) showed an overall coregistration accuracy of 0.99 ± 0.63 mm (explained in further detail in Supplementary material S 3.1, Supplementary Fig. 4) when comparing the anatomical landmarks between PET and MR on the basis of 13 profile lines. Highly perfused areas were observed in the cortical and thalamic regions of the brain and in the region of the olfactory system. Excellent consistency in the coregistration between the anatomical MR data and

Discussion

CBF increases during stimulation revealed by PET imaging were in the lower range (9.7 ± 4.8%) compared to the values reported in the literature, which ranged from 12 ± 2% (Martin et al., 2006) to 47 ± 29% (Weber et al., 2003) and are further discussed in the Supplementary material (S 4.1). Deviations in the CBF change from these previous studies may be explained by anesthetic effects and differences in the stimulation method used. Because one of the aims of the study was to establish a protocol for

Conclusions

It was shown that somatosensory brain activation in rats can be studied using noninvasive PET imaging with [15O]H2O particularly in combination with isoflurane anesthesia, which makes longitudinal experiments in the same animals feasible. Alterations of brain activation can be measured in fast succession enabling a practical stimulation paradigm within one imaging session using the short half-life CBF tracer [15O]H2O. The presented [15O]H2O-PET imaging protocol is not only useful for whisker

Acknowledgments

We thank Anke Stahlschmitt, Maren Koenig, Mareike Lehnhoff and Funda Cay for their excellent technical assistance. This study was supported by the German Research Foundation (DFG) Grant PI 771-1/1, the Werner Siemens-Foundation and the Wilhelm Schuler-Foundation. This study is also part of the PhD thesis of Hans F. Wehrl.

Conflict of interest

B.J.P. receives grant/research support from AstraZeneca, Bayer Healthcare, Boehringer-Ingelheim, Bruker, Oncodesign, Merck, Siemens and the

References (59)

  • F. Lauwers et al.

    Morphometry of the human cerebral cortex microcirculation: general characteristics and space-related profiles

    Neuroimage

    (2008)
  • M. Luijten et al.

    Neurobiological substrate of smoking-related attentional bias

    Neuroimage

    (2011)
  • C. Martin et al.

    Investigating neural-hemodynamic coupling and the hemodynamic response function in the awake rat

    Neuroimage

    (2006)
  • T. Ose et al.

    Quantification of regional cerebral blood flow in rats using an arteriovenous shunt and micro-PET

    Nucl. Med. Biol.

    (2012)
  • W. Schwindt et al.

    Activated areas found by BOLD, CBF, CBV and changes in CMRO2 during somatosensory stimulation do not co-localize in rat cortex

    International Congress Series

    (2002)
  • Y.Y. Shih et al.

    Ultra high-resolution fMRI and electrophysiology of the rat primary somatosensory cortex

    Neuroimage

    (2013)
  • R.G. Shulman et al.

    A BOLD search for baseline

    Neuroimage

    (2007)
  • G. Ambach et al.

    Spatial correlation between sensory regions and the drainage fields of pial veins in rat cerebral cortex

    Exp. Brain Res.

    (1986)
  • P.A. Bandettini et al.

    Spin-echo and gradient-echo EPI of human brain activation using BOLD contrast: a comparative study at 1.5 T

    NMR Biomed.

    (1994)
  • M. Brett et al.

    Region of interest analysis using an SPM toolbox

  • R.B. Buxton

    Introduction to Functional Magnetic Resonance Imaging: Principles and Techniques

    (2009)
  • R.B. Buxton

    Interpreting oxygenation-based neuroimaging signals: the importance and the challenge of understanding brain oxygen metabolism

    Front. Neuroenerg.

    (2010)
  • C. Casteels et al.

    Construction and evaluation of multitracer small-animal PET probabilistic atlases for voxel-based functional mapping of the rat brain

    J. Nucl. Med.

    (2006)
  • S.R. Cherry et al.

    Improved signal-to-noise in PET activation studies using switched paradigms

    J. Nucl. Med.

    (1995)
  • J. Defelipe

    The evolution of the brain, the human nature of cortical circuits, and intellectual creativity

    Front. Neuroanat.

    (2011)
  • R.E. Fish et al.

    Anesthesia and Analgesia in Laboratory Animals

    (2008)
  • P.T. Fox et al.

    Noninvasive functional brain mapping by change-distribution analysis of averaged PET images of H215O tissue activity

    J. Nucl. Med.

    (1989)
  • P.T. Fox et al.

    Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects

    Proc. Natl. Acad. Sci. U. S. A.

    (1986)
  • R.S. Frackowiak et al.

    Functional neuroanatomy of the human brain: positron emission tomography—a new neuroanatomical technique

    J. Anat.

    (1994)
  • Cited by (17)

    • Digital autoradiography for efficient functional imaging without anesthesia in experimental animals: Reversing phencyclidine-induced functional alterations using clozapine

      2020, Progress in Neuro-Psychopharmacology and Biological Psychiatry
      Citation Excerpt :

      Alternative metabolism-reflecting radiolabels, such as 15O, can also be utilized to assess dynamic changes of limited temporal nature (Sheppard et al., 1983; Carter et al., 1997). Longitudinal studies with 15O have been demonstrated feasible in small animals (Wehrl et al., 2014). However, as discussed in this manuscript, most preclinical methods utilizing PET detection are impeded by the requirement of anesthesia or restraint.

    • Future image acquisition trends for PET/MRI

      2015, Seminars in Nuclear Medicine
      Citation Excerpt :

      Hybrid PET/MRI devices offer PET and MRI data with the patient in the same position and almost simultaneous examination time.2 The MRI scanner from a PET/MRI hybrid device exhibits nearly identical image quality compared with their single MRI scanner counterparts, even for advanced functional imaging techniques such as diffusion tensor imaging,3 arterial spin labeling,4 or functional MRI of the brain.5 With the combination of PET and MRI information, a deeper understanding of pathophysiological processes may be expected owing to the elucidation of different functional and metabolic aspects of the disease (Fig. 1).

    View all citing articles on Scopus
    View full text