Hot Topics

Neuropsychopharmacology Reviews (2015) 40, 239–240; doi:10.1038/npp.2014.233

DREAMM: A Biobehavioral Imaging Methodology for Dynamic In Vivo Whole-Brain Mapping of Cell Type-Specific Functional Networks

Michael Michaelides1,2 and Yasmin L Hurd1,2

  1. 1Fishberg Department of Neuroscience, Friedman Brain Institute
  2. 2Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY, USA

Correspondence: Yasmin L Hurd, E-mail:

The human brain contains ~100 billion anatomically and neurochemically heterogeneous neurons that form 100 trillion brain-wide connections giving rise to a complex network whose interconnected activity regulates behavior. A major challenge in neuroscience has been to decipher the discrete functional connectivity underlying normal or pathological behavior.

Until recently, the primary methods for determining whole-brain functional activity as a consequence of discrete manipulation of specific cell populations was to study indirect markers such as c-Fos immunoreactivity in postmortem sections cut throughout the brain, or via in vivo blood oxygenation level-dependent measures using functional magnetic resonance imaging (fMRI), which requires experimental subjects to be fully immobilized. In an effort to overcome such limitations, we recently described the development and application of a new biobehavioral imaging methodology termed DREADD-assisted metabolic mapping (DREAMM) (Anderson et al, 2013; Michaelides et al, 2013). DREAMM leverages the ability of designer receptor exclusively activated by designer drug (DREADD) chemogenetic technology, which enables remote, in vivo activation or inhibition of distinct sets of neurons that result in discrete perturbations in brain activity and behavior (Armbruster et al, 2007), along with the unique capacity of positron emission tomography (PET) and [18F]fluoro-2-deoxyglucose (FDG) for non-invasive, whole-brain, dynamic functional imaging in behaving humans and laboratory animals (Michaelides et al, 2013; Schulz et al, 2011; Villien et al, 2014). DREAMM is thus the first technique that provides simultaneously quantitative, dynamic, whole-brain and regionally unbiased, cell type-specific functional circuit mapping captured during the awake, freely moving state.

DREAMM affords the unique flexibility of combining multiple strategies—molecular, viral, transgenic, and pharmacological (inert ligands that specifically target designer receptors)—to probe, in a mechanistic and unbiased manner, whole-brain functional networks and concomitant behaviors recruited upon stimulation or inhibition of a region- and neurochemically-specific cell type. In a typical DREAMM experiment, subjects are first exposed to a genetic engineering manipulation (eg, viral vector-based or transgenic model), resulting in regional or global expression of stimulatory or inhibitory DREADDs in targeted cell types. Subjects then undergo chemogenetic DREADD modulation by administering an inert ligand such as clozapine-N-oxide that specifically activates the DREADD and is combined with an FDG-PET behavioral imaging strategy. Either a ‘snapshot’ of whole-brain metabolic activity can be obtained during the DREADD-activated freely moving condition prior to imaging in a conventional PET scanner (Michaelides et al, 2013) or using head-mounted mobile PET scanner technology whole-brain metabolic activity is simultaneously measured during behavioral monitoring (Schulz et al, 2011).

DREAMM is currently the only imaging technology that relies on a direct and well-understood process of cellular function (ie, glucose utilization) as its functional output. Moreover, DREAMM is not limited to FDG, and can be implemented along with a variety of other PET probes (even in the same subject) that target alternative/complementary cellular processes such as neurotransmitter dynamics, receptor signaling, and enzymatic processes, thereby allowing investigation not only of the global long-range functional networks involved in cell type-specific behaviors, but also of the underlying neurochemistry. Another unique advantage of DREAMM is its potential to be performed completely non-invasively via the use of transgenic animal models with cell type-specific DREADD expression. Finally, the non-invasive nature of DREAMM when capturing whole-brain in vivo functional activity allows for longitudinal assessments in the same subject and for it to be combined with a variety of model- and disease-specific behavioral paradigms (Figure 1).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact or the author

Sagittal (a) and horizontal (b) views of a three-dimensional image reconstruction of DREAMM showing functional activation in the extended amygdala (ExA) and weaker activation in cingulate cortex (CG) and olfactory tubercle (OT) regions after remote chemicogenetic activation (hM4Di) of prodynorphin-expressing periamygdaloid cortex neurons in behaving rats.

Full figure and legend (67K)Download PowerPoint slide (208 KB)

To date, DREAMM has been used to map novel whole-brain functional activity patterns to changes in motor behavior induced by remote modulation of neighboring, yet heterogeneous basal ganglia pathways (eg, ventral striatonigral and striatopallidal medium spiny neurons). This approach using viral vectors with engineered DREADD receptors under the control of promoters specifically expressed in different striatal pathways illustrated the distinct recruitment of discrete whole-brain functional anatomical signatures and motor behavior upon remote activation of each striatal network (Michaelides et al, 2013). DREAMM has also been used to reverse translate the functional relevance of deficits in neuronal populations that have been virtually unstudied, such as periamygdaloid cortex prodynorphin neurons, in which disturbances were phenotypic of human heroin abusers and major depressive subjects. Novel insights obtained by DREAMM, in translational rodent models, revealed that remote inhibition of periamygdaloid prodynorphin neuronal activity resulted not only in negative affective behavioral regulation, but also in a marked, selective functional activation of the extended amygdala circuit, a brain network well linked to negative affect (Anderson et al, 2013) (Figure 1).

In addition to experimental rodents, DREAMM is now being used in non-human primates and has the potential to be implemented in the future in humans (pending advances in gene-therapy safety/efficacy currently under development). The clinical relevance and translational therapeutics potential of DREAMM holds promise as an application for repeated, longitudinal, non-invasive assessment and monitoring of disease/therapy progression in patients undergoing DREADD-based neuromodulatory therapy. Furthermore, DREAMM would also have the capacity to be utilized alongside advanced cellular genetic engineering approaches such as combined DREADD and cell replacement therapeutics (Dell'Anno et al, 2014) to assess the efficient functional integration of such transplanted cells within neuronal networks.

In sum, DREAMM holds significant promise as a powerful, reverse-engineering strategy to visualize in vivo-specific neuronal ensemble networks associated with normal and pathologic behavior and to thus enhance knowledge about distinct neuronal circuits relevant to neuropsychiatric disorders.



The authors declare that the work was funded by grants from NIDA (DA015446, DA023214, DA030359). MM was supported by the NIDA Postdoctoral Training Program at Icahn School of Medicine at Mount Sinai (DA007135). MM has received compensation from Metis Laboratories and owns stock in the company. YLH declares no potential conflict of interest.



  1. Anderson SA, Michaelides M, Zarnegar P, Ren Y, Fagergren P, Thanos PK et al (2013). Impaired periamygdaloid-cortex prodynorphin is characteristic of opiate addiction and depression. J Clin Invest 123: 5334–5341. | Article | PubMed | ISI |
  2. Armbruster BN, Li X, Pausch MH, Herlitze S, Roth BL (2007). Evolving the lock to fit the key to create a family of G protein-coupled receptors potently activated by an inert ligand. Proc Natl Acad Sci USA 104: 5163–5168. | Article | PubMed | CAS |
  3. Dell'Anno MT, Caiazzo M, Leo D, Dvoretskova E, Medrihan L, Colasante G et al (2014). Remote control of induced dopaminergic neurons in parkinsonian rats. J Clin Invest 124: 3215–3229. | Article | PubMed | ISI |
  4. Michaelides M, Anderson SA, Ananth M, Smirnov D, Thanos PK, Neumaier JF et al (2013). Whole-brain circuit dissection in free-moving animals reveals cell-specific mesocorticolimbic networks. J Clin Invest 123: 5342–5350. | Article | PubMed | ISI |
  5. Schulz D, Southekal S, Junnarkar SS, Pratte JF, Purschke ML, Stoll SP et al (2011). Simultaneous assessment of rodent behavior and neurochemistry using a miniature positron emission tomograph. Nat Methods 8: 347–352. | Article | PubMed | ISI | CAS |
  6. Villien M, Wey HY, Mandeville JB, Catana C, Polimeni JR, Sander CY et al (2014). Dynamic functional imaging of brain glucose utilization using fPET-FDG. Neuroimage 100C: 192–199.

Extra navigation