Highlights d Chemo-connectomics combines chemogenetics (DREADDs) with resting-state fMRI d Locus coeruleus (LC) activation rapidly increases brain-wide functional connectivity d Connectivity changes correlate positively with adrenergic receptor distribution d LC activation shifts large-scale network connectivity toward salience processing

To study brain function, preclinical research heavily relies on animal monitoring and the subsequent analyses of behavior. Commercial platforms have enabled semi high-throughput behavioral analyses by automating animal tracking, yet they poorly recognize ethologically relevant behaviors and lack the flexibility to be employed in variable testing environments. Critical advances based on deep-learning and machine vision over the last couple of years now enable markerless tracking of individual body parts of freely moving rodents with high precision. Here, we compare the performance of commercially available platforms (EthoVision XT14, Noldus; TSE Multi-Conditioning System, TSE Systems) to cross-verified human annotation. We provide a set of videos—carefully annotated by several human raters—of three widely used behavioral tests (open field test, elevated plus maze, forced swim test). Using these data, we then deployed the pose estimation software DeepLabCut to extract skeletal mouse representations. Using simple post-analyses, we were able to track animals based on their skeletal representation in a range of classic behavioral tests at similar or greater accuracy than commercial behavioral tracking systems. We then developed supervised machine learning classifiers that integrate the skeletal representation with the manual annotations. This new combined approach allows us to score ethologically relevant behaviors with similar accuracy to humans, the current gold standard, while outperforming commercial solutions. Finally, we show that the resulting machine learning approach eliminates variation both within and between human annotators. In summary, our approach helps to improve the quality and accuracy of behavioral data, while outperforming commercial systems at a fraction of the cost.

The locus coeruleus (LC) supplies norepinephrine (NE) to the entire forebrain, regulates many fundamental brain functions, and is implicated in several neuropsychiatric diseases. Although selective manipulation of the LC is not possible in humans, studies have suggested that strong LC activation might shift network connectivity to favor salience processing. To test this hypothesis, we use a mouse model to study the impact of LC stimulation on large-scale functional connectivity by combining chemogenetic activation of the LC with resting-state fMRI, an approach we term "chemo-connectomics". LC activation rapidly interrupts ongoing behavior and strongly increases brain-wide connectivity, with the most profound effects in the salience and amygdala networks. We reveal a direct correlation between functional connectivity changes and transcript levels of alpha-1, alpha-2, and beta-1 adrenoceptors across the brain, and a positive correlation between NE turnover and functional connectivity within select brain regions. These results represent the first brain-wide functional connectivity mapping in response to LC activation, and demonstrate a causal link between receptor expression, brain states and functionally connected large-scale networks at rest. We propose that these changes in large-scale network connectivity are critical for optimizing neural processing in the context of increased vigilance and threat detection. RESULTSTo selectively target the LC, we used transgenic mice that express codon-improved Crerecombinase (iCre) under the dopamine-beta-hydroxylase (DBH) promoter (DBH-iCre mice, Figure 1A) 33 . We stereotactically delivered floxed excitatory DREADDs 34 (AAV5-hSyn-DIO-hM3Dq-mCherry; hM3Dq-mCh) or a control virus (AAV5-hSyn-DIO-mCherry; mCh) to the LC, thus restricting virus expression to DBH-positive noradrenergic neurons of the LC (Figure 1B). We assessed successful LC activation using pupillometry, a highly sensitive and clinically relevant readout of LC activation [35][36][37] . After two minutes of baseline recording under light isoflurane anesthesia, we activated LC neurons by administering the potent DREADD activator clozapine at an ultra-low dose (0.03 mg/kg, i.p., Figure 1C) 38 . Within a minute of clozapine injection, we observed a strong increase in pupil diameter in the hM3Dq-mCh group, while pupil diameter of mCh mice did not change in response to clozapine injection and remained stable throughout the 10-minute recording session (Figure 1D-F). To show that our LC activation protocol is behaviorally relevant, we subjected mice to an open field test (OFT) immediately after clozapine injection and recorded their behavior for 30 minutes. In comparison to mCh mice, clozapine injection had profound effects on the behavior of hM3Dq-mCh mice. Several minutes after clozapine administration, hM3Dq-mCh mice showed strongly suppressed locomotor activity ( Figure 1G, H), spent less time in the (more aversive) center of the open field (Figure 1I, J), performed less activity-related supported rears (Figure 1...