Counting cells in fluorescent microscopy is a tedious, time-consuming task that researchers have to accomplish to assess the effects of different experimental conditions on biological structures of interest. Although such objects are generally easy to identify, the process of manually annotating cells is sometimes subject to fatigue errors and suffers from arbitrariness due to the operator’s interpretation of the borderline cases. We propose a Deep Learning approach that exploits a fully-convolutional network in a binary segmentation fashion to localize the objects of interest. Counts are then retrieved as the number of detected items. Specifically, we introduce a Unet-like architecture, cell ResUnet (c-ResUnet), and compare its performance against 3 similar architectures. In addition, we evaluate through ablation studies the impact of two design choices, (i) artifacts oversampling and (ii) weight maps that penalize the errors on cells boundaries increasingly with overcrowding. In summary, the c-ResUnet outperforms the competitors with respect to both detection and counting metrics (respectively, $$F_1$$F 1 score = 0.81 and MAE = 3.09). Also, the introduction of weight maps contribute to enhance performances, especially in presence of clumping cells, artifacts and confounding biological structures. Posterior qualitative assessment by domain experts corroborates previous results, suggesting human-level performance inasmuch even erroneous predictions seem to fall within the limits of operator interpretation. Finally, we release the pre-trained model and the annotated dataset to foster research in this and related fields.

Hibernation has been proposed as a tool for human space travel. In recent years, a procedure to induce a metabolic state known as “synthetic torpor” in non-hibernating mammals was successfully developed. Synthetic torpor may not only be an efficient method to spare resources and reduce psychological problems in long-term exploratory-class missions, but may also represent a countermeasure against cosmic rays. Here we show the preliminary results from an experiment in rats exposed to ionizing radiation in normothermic conditions or synthetic torpor. Animals were irradiated with 3 Gy X-rays and organs were collected 4 h after exposure. Histological analysis of liver and testicle showed a reduced toxicity in animals irradiated in torpor compared to controls irradiated at normal temperature and metabolic activity. The expression of ataxia telangiectasia mutated (ATM) in the liver was significantly downregulated in the group of animal in synthetic torpor. In the testicle, more genes involved in the DNA damage signaling were downregulated during synthetic torpor. These data show for the first time that synthetic torpor is a radioprotector in non-hibernators, similarly to natural torpor in hibernating animals. Synthetic torpor can be an effective strategy to protect humans during long term space exploration of the solar system.

Tau protein is of primary importance for many physiological processes in neurons, where it affects the dynamics of the microtubule system. When hyperphosphorylated (PP-Tau), Tau monomers detach from microtubules and tend to aggregate firstly in oligomers, and then in neurofibrillary tangles, as it occurs in a group of neurodegenerative disorders named thauopathies. A hypothermia-related accumulation of PP-Tau, which is quickly reversed after the return to normothermia, has been shown to occur in the brain of hibernators during torpor. Since, recently, in our lab, a hypothermic torpor-like condition (synthetic torpor, ST) was pharmacologically induced in the rat, a non-hibernator, the aim of the present work was to assess whether ST can lead to a reversible PP-Tau accumulation in the rat brain. PP-Tau was immunohistochemically assessed by staining for AT8 (phosphorylated Tau) and Tau-1 (non-phosphorylated Tau) in 19 brain structures, which were chosen mostly due to their involvement in the regulation of autonomic and cognitive functions in relation to behavioral states. During ST, AT8 staining was strongly expressed throughout the brain, while Tau-1 staining was reduced compared to control conditions. During the following recovery period, AT8 staining progressively reduced close to zero after 6 h from ST. However, Tau-1 staining remained low even after 38 h from ST. Thus, overall, these results show that ST induced an accumulation of PP-Tau that was, apparently, only partially reversed to normal during the recovery period. While the accumulation of PP-Tau may only depend on the physicochemical characteristics of the enzymes regulating Tau phosphorylation, the reverse process of dephosphorylation should be actively regulated, also in non-hibernators. In conclusion, in this work a reversible and widespread PP-Tau accumulation has been induced through a procedure that leads a non-hibernator to a degree of reversible hypothermia, which is comparable to that observed in hibernators. Therefore, the physiological mechanism involved in this process can sustain an adaptive neuronal response to extreme conditions, which may however lead to neurodegeneration when particular intensities and durations are exceeded.

Tau is a key protein in neurons, where it affects the dynamics of the microtubule system. The hyperphosphorylation of Tau (PP-Tau) commonly leads to the formation of neurofibrillary tangles, as it occurs in tauopathies, a group of neurodegenerative diseases, including Alzheimer's. Hypothermia-related accumulation of PP-Tau has been described in hibernators and during synthetic torpor (ST), a torpor-like condition that has been induced in rats, a non-hibernating species. Remarkably, in ST PP-Tau is reversible and Tau de-phosphorylates within a few hours following the torpor bout, apparently not evolving into pathology. These observations have been limited to the brain, but in animal models of tauopathies, PP-Tau accumulation also appears to occur in the spinal cord (SpCo). The aim of the present work was to assess whether ST leads to PP-Tau accumulation in the SpCo and whether this process is reversible. Immunofluorescence (IF) for AT8 (to assess PP-Tau) and Tau-1 (non-phosphorylated Tau) was carried out on SpCo coronal sections. AT8-IF was clearly expressed in the dorsal horns (DH) during ST, while in the ventral horns (VH) no staining was observed. The AT8-IF completely disappeared after 6 h from the return to euthermia. Tau-1-IF disappeared in both DH and VH during ST, returning to normal levels during recovery. To shed light on the cellular process underlying the PP-Tau pattern observed, the inhibited form of the glycogen-synthase kinase 3β (the main kinase acting on Tau) was assessed using IF: VH (i.e., in motor neurons) were highly stained mainly during ST, while in DH there was no staining. Since tauopathies are also related to neuroinflammation, microglia activation was also assessed through morphometric analyses, but no ST-induced microglia activation was found in the SpCo. Taken together, the present results show that, in the DH of SpCo, ST induces a reversible accumulation of PP-Tau. Since during ST there is no motor activity, the lack of AT8-IF in VH may result from an activity-related process at a cellular level. Thus, ST demonstrates a newly-described physiological mechanism that is able to resolve the accumulation of PP-Tau and apparently avoid the neurodegenerative outcome.