Temperature‐Dependent Relationship of Autophagy and Apoptotic Signaling During Cold‐Water Immersion in Young and Older Males

Autophagy is a crucial cytoprotective mechanism preventing the accumulation of cellular damage, especially during external stimuli such as cold exposure. Older adults poorly tolerate cold exposure and age‐related impairments in autophagy may contribute to the associated reductions in cold tolerance. The purpose of this investigation is to evaluate the effect of different intensities of in vivo cold‐water immersion and in vitro cold exposure on autophagic and apoptotic signaling in young and older males. Peripheral blood mononuclear cells (PBMCs) are isolated at baseline, end‐cold exposure, and after 3 h of thermoneutral recovery. Additionally, PBMCs are treated with rapamycin and bafilomycin prior to in vitro cold exposure equivalent to in vivo core temperatures (35–37 °C). Proteins associated with autophagy, apoptosis, the heat shock response, and inflammation are analyzed via Western blotting. Moderate cold stress (0.5 °C decrease in core temperature) increased autophagic and heat shock protein activity while high cold stress (1.0 °C decrease in core temperature) augmented apoptosis in young males. In older males, minimal autophagic activation during both cold‐water exposures are associated with increased apoptotic and inflammatory proteins. Although in vitro cold exposure confirmed age‐related dysfunction in autophagy, rapamycin‐induced stimulation of autophagic proteins underlie the potential to reverse age‐related vulnerability to cold exposure.


Introduction
It is well established that aging alters the ability to tolerate cold exposure. [1]This is best demonstrated by the inability of older DOI: 10.1002/adbi.202300560adults to maintain core temperature at or near resting levels (≈37 °C) when exposed to cold ambient conditions considered to be compensable (i.e., sufficient heat conservation to preserve core temperature) in young adults. [2]These age-associated decrements in thermoregulation are attributed to reductions in peripheral vascular function, leading to higher peripheral skin temperature and therefore greater heat loss. [3]Consequently, prolonged durations of cold exposure combined with impairments in heat conserving mechanisms can translate to dangerous decreases in core temperature during a cold exposure that can threaten the health of vital organs, especially as core temperatures approach and exceed the threshold of hypothermia (35 °C). [4]Although there are well defined age-related decrements in the physiological mechanisms governing the regulation of heat conserving response (i.e., peripheral skin vasoconstriction) that increase the risk of cold-induced injuries (including hypothermia), [3] there exists a paucity of information regarding the impact of age on cellular function during acute cold exposure.
At the cellular level, cold stress causes profound perturbations to cellular homeostasis leading to the disruption of normal cell function, in part, attributed to cold-induced protein denaturation stemming from changes in the polarity properties of water that occurs with cooling. [5]These disruptions in cellular homeostasis that lead to an accumulation of damaged proteins and organelles can develop into cellular and eventually tissue death that threatens human life. [6]This is further compounded in older adults, who experience a reduction in proteostatic mechanisms that can lead to greater accumulation of partially folded, misfolded, and metastable protein aggregates, [7] thereby reducing the ability of aged cells to tolerate exposure to cold-induced stress.In response to cold exposure, cellular protective mechanisms, including the vital survival mechanism of macroautophagy (herein referred to as autophagy), are initiated to address the accumulation of toxic protein aggregates through autophagosomal degradation of damaged proteins and organelles that are subsequently broken down into their basic constituents. [6]The autophagic response can be initiated via a multitude of stimuli including external stressors (e.g., environmental) [8] or physiological stressors (e.g., nutrient deprivation or metabolic stress). [9]As a downstream mechanism to energy sensing pathways such as AMPK and in turn mTOR inhibition, [10] autophagy is an integral component of the cellular quality control network and works in tandem with pathways such as the heat shock response (HSR) to maintain the balance of the cellular proteome using heat shock proteins (HSPs) to suppress inflammation [11] and subsequently, apoptotic cell death. [11,12]The process of autophagy is especially crucial during exposure to environmental thermal challenges such as cold exposure (and the successive re-warming) to counteract rates of apoptosis and necrosis as a result of increased cold-induced protein denaturation, reduced global protein synthesis, and cellular depolarization by an influx of calcium. [13]The vital role of autophagy is extended during the re-warming process where stress granules composed of protein aggregates and untranslated mRNA are potentially broken down by autophagy to facilitate rapid recovery. [14]However, it is unknown how the age-related decline in autophagy [15] may impede responses to cold exposure and the subsequent recovery process.
To date, the interaction between cellular mechanisms of cell survival (i.e., autophagy and HSPs) and cell death systems (i.e., apoptosis) have yet to be evaluated during exposure to physiological relevant cold conditions in humans.Moreover, the extent of dysfunctional autophagy during cold conditions is unknown when considering factors such as aging where autophagy is known to be reduced. [15]Addressing this critical knowledge gap may provide the first steps in identifying the underlying cellular mechanisms associated with age-related vulnerability to cold exposure and may help develop strategies (i.e., through the use of autophagic stimuli) to potentially reverse age-related reductions in autophagy.Thus, the primary purpose of this investigation was to establish the body temperature-dependent effects of in vivo cold exposure in peripheral blood mononuclear cells (PBMCs) from young (24 ± 3 years) and older (61 ± 3 years) males during whole-body cold-water immersion (indexed by a decrease in resting esophageal temperature by 0.5 and 1.0 °C).To better understand the mechanistic flexibility of autophagy during cold exposure in the absence of shivering-induced metabolic stress, the secondary purpose of this investigation aimed to establish the temperature-dependent effect (35-37 °C) of the autophagic response during incremental in vitro cold exposure.Further, to evaluate the potential to reverse age-related decrements in autophagy during cold exposure, in vitro cold exposures were performed with and without treatments of a known autophagy initiator (rapamycin).It was hypothesized that moderate decreases in core temperature (as defined by a reduction in core temperature 0.5 °C below resting levels) during cold-water immersion will induce autophagy, while more robust decreases in core temperature (as defined by a reduction in core temperature 1.0 °C below resting levels) will elicit a shift toward apoptotic responses.It was also hypothesized that the established age-related decline in basal autophagic function in older adults [15] would be associated with an attenuated response in autophagy during cold exposure and a heavier reliance on apoptotic pathways relative to their younger counterparts, which will be alleviated with treatments of rapamycin.

In Vivo Cold-Water Immersion
To investigate the impact of age on autophagic activity during cold exposure, we first aimed to utilize cold-water immersion to elicit a reduction in core temperature (indexed by esophageal temperature) by 0.5 °C (moderate cold stress) and 1.0 °C (high cold stress).By design, this was achieved and maintained for 30 min in both young and older males, in which core temperature was significantly different between conditions (all p < 0.001; Figure 1A and  B).No differences were observed between baseline core temperature following a 30 min habituation period between the young and older males (p = 0.605; Figure 1C and D).Between the two conditions, the desired endpoints were attained with marginally different water temperatures (mean ± standard deviation (SD): 20.1 ± 3.9 °C and 18.7 ± 4.0 °C, respectively; p < 0.001).However, the total immersion times for the moderate and high cold stress conditions (42 ± 22 min and 50 ± 21 min, respectively; p = 0.217) and the time to rewarm in 39 °C water (46 ± 12 min and 48 ± 13 min, respectively; p = 0.458) did not differ.

Autophagic Responses are Attenuated in Older Males During Cold-Water Immersion
Autophagic function was analyzed via changes in protein expression of several commonly used markers (microtubule-associated protein 1 light chain 3 [LC3], sequestosome-1 [p62], unc-51-like autophagy-activating kinase 1 [ULK1], phospho-ULK1 [pULK1], and beclin-2).An increase in LC3-II levels along with a decrease in p62 was interpreted as an increase in autophagy with increases in total ULK1 (herein referred to as ULK1), pULK1, and beclin-2 interpretated as an increase in autophagic induction. [6]Protein quantifications are reported as a relative quantity (RQ) to the baseline of the respective day.During the in vivo moderate cold stress condition, LC3-II increased from baseline immediately post-cold exposure and remained elevated 3 h following cold exposure in the young (1.45 ± 0.16 RQ and 1.72 ± 0.22 RQ, respectively; both p ≤ 0.017; Figure 2A), but not the older males (0.99 ± 0.09 RQ and 1.05 ± 0.18 RQ, respectively; both p ≥ 0.747; Figure 2B).Additionally, LC3-II was significantly lower in the older compared to young males (Figure 1B) immediately postcold exposure (p = 0.034) and following 3 h of recovery (p = 0.004).However, during the high cold stress condition, LC3-II did not increase at the end of cold exposure or following the 3 h recovery period in neither the young (0.88 ± 0.09 RQ and 0.88 ± 0.11 RQ, respectively; both p ≥ 0.495) nor older males (0.76 ± 0.12 RQ and 0.76 ± 0.16 RQ, respectively; both p ≥ 0.117).Further, there were significant increases of LC3-II/LC3-I during cold-water immersion in the young (main effect: p = 0.038; Figure 2C) but not the older males (main effect: p = 0.697; Figure 2D), with significantly lower LC3-II/I in older males than young males during the high cold stress condition (p ≤ 0.038; Figure 2D).
In the young males, a significant reduction in p62 was observed following cold exposure (main effect: p = 0.018; Figure 2E) irrespective of condition.Conversely, a significant accumulation in p62 occurred in the older males following cold exposure (main effect: p = 0.011; Figure 2F) in both conditions that was also significantly higher than in young males (all p ≤ 0.012; Figure 2F).Further, beclin-2 significantly increased following cold exposure (1.53 ± 0.13 RQ, p < 0.001) and recovery (1.40 ± 0.13 RQ, p < 0.001) during the moderate cold stress condition in young males, but not during the high cold stress condition (0.85 ± 0.08 RQ and 0.92 ± 0.09 RQ, respectively; both p ≥ 0.140) (Figure 2G).Similarly, ULK1 significantly increased following cold exposure (p = 0.004) and recovery (p < 0.001) only in the young males during the moderate cold stress condition, although, there were no significant differences in pULK1 either group (main effect: both p ≥ 0.222) (Table 1).

Apoptotic Signaling is Elevated in Older Males and High Cold Stress
Apoptotic signaling was assessed through changes in caspase-3.Specifically, quantification of the cleavage of pro-caspase-3 (cleaved-caspase-3) was interpreted as the execution of apoptosis.Following cold exposure in the moderate cold stress condition, cleaved-caspase-3 significantly increased in the older (1.77 ± 0.12 RQ, p < 0.001; Figure 3A), but not the young males (1.07 ± 0.10 RQ, p = 0.620; Figure 3B).However, following exposure to the high cold stress condition, cleaved-caspase-3 significantly increased in both the young (1.62 ± 0.14 RQ, p < 0.001) and older males (2.51 ± 0.21, p < 0.001).Additionally, cleaved-caspase-3 remained elevated in the older males during the high cold stress condition recovery (1.71 ± 0.12 RQ, p = 0.002).Despite increases in cleaved-caspase-3 in both young and older males following the high cold stress condition, older males had significantly higher cleaved-caspase-3 than young males following cold exposure and recovery in both conditions (all p ≤ 0.033; Figure 3B).In the young males, cleaved/pro-caspase-3 (Figure 3C) was significantly higher in the high compared to the moderate cold stress condition (main effect: p = 0.008).In the older males, the ratio of cleaved/pro-caspase-3 (Figure 3D) was significantly higher following the cold exposure of the moderate (1.42 ± 0.11 RQ, p = 0.048) that was significantly higher than the young males (p < 0.001), in addition to an increase in cleaved/pro-caspase-3 during the high cold stress condition (2.20 ± 0.31 RQ, p < 0.001).

The Heat Shock and Inflammatory Response to Cold-Water Exposures
Changes in HSP70 and HSP90 protein expression were representative in the HSR, while the acute inflammatory response were interpreted with changes in tumor necrosis factor-alpha (TNF-) and interleukin-6 (IL-6) protein expression.Given autophagy is exclusively an intracellular mechanism, [12] intracellular protein expression of the HSR and inflammatory response was assessed.Following cold exposure and recovery in the moderate cold stress condition, both HSP90 (both p ≤ 0.044) and HSP70 (both p ≤ 0.042) significantly increased in the young males (Table 1).No significant changes in HSP90 or HSP70 were observed in the older males in either cold-water condition (Table 1).No significant changes in IL-6 were observed in the young males (main effect: p = 0.578), however, a significant increase in response to cold-water immersion was observed in the older males (main effect: p = 0.042; Table 1).Further, there was a significant increase in TNF- following cold-water immersion in the older males (main effect: p = 0.030), while TNF- in the young  i) *Indicates a significant increase (all p ≤ 0.044) from baseline (BL); ii) # Indicates a significant increase by condition (all p ≤ 0.030) in older males; iii)  Indicates a significant change by time (all p ≤ 0.042) in older males; iv) § Indicates significantly higher in older than young males (main effect: p = 0.034) during the moderate cold stress condition.
males only increased following cold exposure (p = 0.015) and recovery (p = 0.043) in the high cold stress condition (Table 1).However, while there were no significant differences in IL-6 between young and older males for either condition (main effect: p ≥ 0.068), TNF- was significantly higher than in older males during the moderate cold stress condition (main effect: p = 0.034; Table 1).

Rapamycin Reverses Age-Induced Autophagic Dysfunction in Cold Conditions
To further investigate if priming autophagic activity can alter the age-related decrements in cytoprotection during cold exposure, we then administered rapamycin treatments to PBMCs during in vitro cooling (37.0, 36.5, 36.0, and 35.0 °C).Additionally, bafilomycin treatments were administered to amplify the resolution of autophagic responses by preventing the degradation of key autophagy markers. [6]Protein quantifications are reported as a relative quantity to the control treatment (dimethylsufoxide; DMSO) of the respective temperature condition.In response to rapamycin and bafilomycin during cooling, autophagic activity increased as depicted by elevated levels of LC3-II from the control during the 37.0 and 36.5 °C conditions with treatments of bafilomycin (2.17 ± 0.21 RQ and 2.15 ± 0.23 RQ, respectively; p ≤ 0.001), rapamycin (1.46 ± 0.15 RQ and 1.26 ± 0.08 RQ, respectively; p ≤ 0.014), and bafilomycin with rapamycin (3.03 ± 0.37 RQ and 2.94 ± 0.41 RQ, respectively; p ≤ 0.001) in the young males (Figure 4A).Additionally in the young males, LC3-II was significantly higher with treatments of bafilomycin with rapamycin than bafilomycin alone during the 37.0 °C (p = 0.022) and 36.5 °C (p = 0.004) conditions.During the 36.0 °C cooling, LC3-II was significantly higher than the control in the young males with treatments of bafilomycin (1.53 ± 0.15 RQ, p = 0.021) and bafilomycin with rapamycin (1.67 ± 0.17 RQ, p = 0.003), but were not significantly different between treatment conditions (p = 0.622).
In the older males, LC3-II significantly increased from the control with bafilomycin (1.28 ± 0.08 RQ, p = 0.009) and bafilomycin with rapamycin (1.59 ± 0.12 RQ, p = 0.001) in the 37.0 °C condition, but only bafilomycin with rapamycin (1.43 ± 0.12 RQ, p = 0.007) was significantly higher than the control in the 36.5 °C condition (Figure 4B).LC3-II was also significantly higher in the 37.0 °C condition with treatments of bafilomycin with rapamycin than only bafilomycin in the older males (p = 0.004).Despite increases in LC3-II with rapamycin and bafilomycin, older males had significantly lower LC3-II than young males with these treatments during the 37.0 °C (p = 0.002 and p = 0.003, respectively) and 36.5 °C (p = 0.006 and p = 0.002, respectively) conditions.Additionally, LC3-II was significantly lower in older males than young males in the 36 °C condition with all treatments (p ≤ 0.045).LC3-II/LC3-I during the 37.0 °C condition was significantly higher than the control with treatments of bafilomycin (1.96 ± 0.18 RQ, p = 0.001), and bafilomycin with rapamycin (2.38 ± 0.25 RQ, p < 0.001) with significantly higher bafilomycin with rapamycin than bafilomycin alone (p = 0.049) in the young males (Figure 4C).Additionally in the young males, LC3-II/LC3-I was significantly higher than the control in the 36.5°Ccondition with treatments of bafilomycin (1.90 ± 0.20 RQ, p = 0.002) and bafilomycin with rapamycin (2.05 ± 0.19 RQ, p < 0.001), and significantly higher than the control in the 36.0 °C condition with bafilomycin and rapamycin (1.65 ± 0.26 RQ, p = 0.034) treatments.No significant differences in LC3-II/LC3-I were observed in the older males (main effect: p = 0.251; Figure 4D).Further, LC3-II/I was significantly lower in older males compared to young males (Figure 4D) during the 37.0 and 36.5 °C with treatments of bafilomycin (both p = 0.001, respectively) and bafilomycin with rapamycin (p = 0.001 and p = 0.002, respectively) as well as during the 36.0 °C condition with treatments of rapamycin (p = 0.001) and bafilomycin with rapamycin (p = 0.045).
In the young males, p62 (Figure 4E) was significantly lower in the 37.0 and 36.5 °C conditions with treatments of bafilomycin (0.78 ± 0.06 RQ, p = 0.007), rapamycin (0.76 ± 0.04 RQ, p < 0.001), and bafilomycin with rapamycin (0.64 ± 0.03 RQ, p < 0.001).Further, p62 with bafilomycin and rapamycin treatments were significantly lower than only bafilomycin in the 37.0 °C condition of the young males (p = 0.041).In the older males, significant elevations in p62 were observed from control (main effect: p = 0.008) and with decreasing temperature (main effect: p = 0.018) (Figure 4F).Additionally, p62 was significantly higher in older males compared to young males during the 36.5 °C condition with all treatments (all p ≤ 0.012; Figure 4F).A significant increase in beclin-2 was observed with cooling (main effect: p = 0.048) in the young males with no significant differences observed in the older males (main effect: p = 0.788; Table 2).In both the young and older males, ULK1 significantly increased from the control condition with treatments of bafilomycin (both p ≤ 0.002), rapamycin (both p ≤ 0.015), and bafilomycin with rapamycin (both p < 0.001) in the 37.0 °C condition, as well as bafilomycin (both p ≤ 0.013) and bafilomycin with rapamycin (both p ≤ 0.005) in the 36.5 °C condition (Table 2).Only the young males had significantly higher ULK1 in the 36.0 °C condition with treatments of bafilomycin (p = 0.017) and bafilomycin with rapamycin (p = 0.037).Further, both the young and older males had significantly higher ULK1 with bafilomycin and rapamycin treatments than the bafilomycin treatments in the 37.0 °C condition (both p ≤ 0.023).pULK1 similarly increased compared to control in the young males during the 37.0 °C condition with treatments of bafilomycin (p = 0.004), rapamycin (p = 0.031), and bafilomycin with rapamycin (p < 0.001), however, only the bafilomycin with rapamycin (p = 0.001) treatment increased in the older males during the 37.0 °C treatment (Table 2).In both the young and older males, bafilomycin with rapamycin treatments were significantly higher than the bafilomycin treatment during the 37.0 °C condition (both p ≤ 0.020).In the 36.5 °C condition, pULK1 from both bafilomycin (p = 0.002) and bafilomycin with rapamycin (p < 0.001) treatments increased compared to control in the young males with no significant differences between the two treatments (p = 0.979) (Table 2).In the older males, pULK1 only increased compared to control in the bafilomycin with rapamycin (p < 0.001) treatment during the 36.5 °C condition, but also increased in the rapamycin (p = 0.012) treatment during the 36.0 °C condition (Table 2).

In Vitro Heat Shock and Inflammatory Responses
In the young males, a significant increase from control was observed in HSP90 (main effect: p = 0.002) (Table 2).No other significant changes were observed in HSP90 in the older males (main effects: both p ≥ 0.107) nor in HSP70 for either the young males (main effects: both p ≥ 0.110) or older (main effects: p ≥ 0.093) (Table 2).Additionally in the young males, a significant increase from control (main effect: p = 0.005) was observed in IL-6 (Table 2).No other significant changes were observed in IL-6 in the older males (main effect: p = 0.107) nor in TNF- for either the young (main effect: p = 0.095) or older males (main effect: p = 0.283) (Table 2).

Discussion
While limited animal and cellular models have sought to identify the autophagic response to cold exposure, there remains a vast knowledge gap regarding autophagic responses to cold exposure in humans, including the underlying mechanisms associated with age-related vulnerability to cold exposure.To establish the cold-intensity relationship of autophagy in the context of aging, the primary objective of the present investigation was to identify the effects of cold exposure on the autophagic response in PBMCs during exposure to moderate and high cold stress (a 0.5 and 1.0 °C reduction in resting core temperature, respectively) via cold-water immersion.As a secondary objective, PBMCs were isolated from age-matched young and older adults to evaluate if autophagic responses can be altered during in vitro cold exposure (37.0, 36.5, 36.0, and 35.0 °C) with rapamycin treatments.In support of our hypothesis, an intensity-dependent transition was observed from autophagic activity with moderate cold stress to favoring apoptotic signaling with high cold stress in young males.Additionally, we showed there is a clear delineation of agerelated autophagic decline via lower LC3-II, an accumulation in p62, and concomitant shift toward apoptotic mechanisms during cold exposure in older compared to young males.The attenuated autophagic responses were partially restored during cold exposure through in vitro treatment of rapamycin in older adults, as evidenced through an associated increase in LC3-II, albeit an increase in apoptotic activity was still observed during exposure to a moderate decrease in temperature (equivalent to moderate cold stress).Taken together, our findings demonstrate for the first time that the age-related decline in basal autophagy [15] is exacerbated during cold exposure with increasing intensities of cold exposure.Further, we provide some of the first insights into pharmacological simulation of autophagy as a potential avenue to enhance cytoprotection during cold exposure in vulnerable older adults who experience age-related declines in autophagy.

The Effect of In Vivo Cold Exposure on the Autophagic Response
Given in vivo cold exposure entails both metabolic and thermal perturbations, cold exposure can elicit profound disruptions to cellular proteostatic mechanisms.Despite a handful of instances where autophagy has been shown to increase in response to cold exposure including in AML12 cells [16] and zebrafish, [17] to the authors' knowledge no studies have examined the human autophagic response during in vivo cold exposure.For the first time in humans, we demonstrate an increase in key autophagy proteins during moderate cold stress, as evidenced by an increase in LC3-II and LC3-II/I in young males (Figure 1A and C), however, this was attenuated with higher intensities of cold stress (Figure 1A and C).Further supporting our observations of elevated autophagy during exposure to a moderate cold stress, we showed a decrease in autophagy cargo protein p62 (which typically declines with increasing autophagy) (Figure 1E), that was paralleled by an increase in autophagic initiation proteins beclin-2 (Figure 1G) and ULK1 (Table 1).Conversely, while high cold stress elicited a decrease in p62 in young males, no further elevations in beclin-2 or ULK1 were observed, suggesting an attenuation of autophagic initiation.Lower autophagic activity during high cold stress may have shifted the cell to an environment favoring apoptotic responses as supported by elevations in apoptotic proteins cleaved-capsase-3 (Figure 2A) and cleaved/pro-caspase-3 (Figure 2C).This phenomenon of increased apoptosis is known to occur with higher intensities and duration of cold exposure as shown in Burkitt lymphoma cells, [18] but until now has yet to be shown in humans.Together, this indicates that there may exist a threshold between the two different levels of cold stress assessed within this investigation in that cellular signaling promotes cytoprotective autophagy at moderate intensities of cold stress, whereas a shift to an apoptotic state occurs as core temperatures approach greater levels of hypothermia as observed during the high cold stress exposure.
Cold exposure elicits a complex stress on the cell with activation of many pathways that potentially influence the autophagic response, including interactions with the acute inflammatory response and HSR.This interconnection is demonstrated through activation of HSP90 which inhibits apoptosis through the suppression of TNF-. [11]In the present investigation, HSP90 increased in response to the moderate, but not high cold stress condition (Table 1).Conversely, TNF- did not change during the moderate cold stress condition but was elevated during the high cold stress condition (Table 1).Therefore, our findings observed in the moderate cold stress condition support the antiinflammatory attributes of autophagy [19] and the HSR. [11]Despite an increase in TNF- during the high cold stress condition suggesting an increase in intracellular inflammation, no changes were observed in IL-6 in young males (Table 1).This may be attributed to the level of cold stress in the present study representing cold stress-inducing reductions in core temperature that are physiologically relevant to humans, as others have observed elevations in both TNF- and IL-6 in the colon of C57BL/6 mice exposed to extreme temperatures of 4 °C for 3 h. [20]nterestingly in the present study, we observed activation of the HSR during moderate cold stress, as represented by elevations in both HSP70 and HSP90 immediately following cold exposure, that were sustained throughout the 3 h recovery (Table 1).Although stressors that upregulate HSPs initially increase autophagy during or immediately following a stress insult (as also shown in the present investigation), HSPs typically upregulate following the cessation of the stressor. [12]This is seen in cold exposure where HSP90 mRNA expression increases following the rewarming of porcine fibroblasts exposed to 4 and 15 °C. [21]Similarly, HSP70 overexpression delays the initiation of starvationinduced autophagy in A549 cells [12] and inhibited autophagy in Drosophila melanogaster following 1 h exposures of 0 to −12 °C. [22]here are, however, instances where autophagic activity concurrently increases with HSP70 as seen during ex vivo heating in PBMCs isolated from young adults [23] and concurrently increases with HSP90 during in vivo vigorous-intensity exercise (70% VO 2max ) with and without environmental heating (40 °C) in young males. [24]This may indicate that the simultaneous expression of HSPs with autophagic activity may be intensity dependent, given we did not observe concurrent activation during www.advanced-bio.com the high cold stress condition (Table 1), although this remains speculative.

The Effect of Aging During In Vivo Cold Exposure on the Autophagic Response
The steady deterioration of autophagic function associated with aging is linked to the reduction of lysosomal proteolytic function to breakdown the material bound within the autolysosome. [25] key indicator of autophagic dysregulation occurs with an absence of change in LC3-II [26] in coordination with an accumulation in the autophagy machinery of p62. [27]In line with this interpretation and contrary to the young males, we did not observe any changes in LC3-II in the older males in either condition (Figure 2B).When paired with a simultaneous increase in p62 (Figure 2F) our findings support the notion of autophagic dysregulation in older males during cold stress.Further, the contrast in age-related autophagic activity in PBMCs is consistent with observations during ex vivo heating. [28]Coinciding with no significant changes in the proteins associated with autophagic regulation, an increase in cleaved-caspase-3 (Figure 2B) and cleaved/pro-caspase-3 were observed immediately post-cold exposure in both the moderate and high cold stress conditions.As such, the premature induction of apoptosis that was not observed in young males in the milder condition of the two cold stress intensities may indicate that autophagy contributes to the established age-related reduction in cold tolerance. [1]n concert with age-related declines in basal autophagy, [15] chronically excessive activation of chaperone-mediated proteins (including HSPs) will reduce their function. [29]The alteration in HSP function presents as changes in the transcriptome (mRNA sequencing) that changes the chaperone network function as seen in aging brains with and without neurodegenerative disease. [30]The present investigation supports this concept of altered chaperone activity as evidenced by no observed changes in HSP90 in response to in vivo cooling which contrasts the significant increase in young males during the moderate cold stress condition (Table 1).Although there was a significant increase in HSP70 from older males with cold-water immersion, there were also elevations in both IL-6 and TNF- (Table 1).The increased inflammatory signaling, however, may be a result of senoinflammation, where low-grade age-associated chronic inflammation may exaggerate the inflammatory response to external stimuli such as cold exposure. [31]

The Effect of Aging During In Vitro Cold Exposure with Bafilomycin and Rapamycin Treatments on the Autophagic Response
To better understand the temperature-dependent relationship in autophagic responses to cold exposure, in vitro cold exposures were used to eliminate compounding stimuli from cold-induced autonomic nervous activity [32] or oxidative stress as a byproduct of shivering muscles, [33] and to assess temperatures not viable for human testing (i.e., sustained hypothermia, 35 °C).Further, treatments of bafilomycin and rapamycin were used to enhance the interpretation of autophagic activity and assess the potential for improvements in age-related autophagic dysfunction during cold exposure in older adults.In the present investigation, autophagy increased with bafilomycin and rapamycin treatments in young males as evident by increases in LC3-II (Figure 4A) and ULK1 (Table 2) at 37.0, 36.5, and 36.0 °C and a decrease in p62 at 37 and 36.5 °C (Figure 4E).Despite the addition of bafilomycin to reduce the degradation of the autophagic machinery for improved interpretation, our findings of decreased p62 in young males and elevated p62 in older males compared to young males (Figure 4C and F) coincide with a similar investigation assessing rapamycin and bafilomycin treatments of PBMCs harvested from young (22 ± 2 years) and older (64 ± 4 years) adults exposed to 37 °C. [34]ogether, this suggests that autophagic responses to in vitro cold exposure equivalent to moderate and high in vivo cold stress may be improved and can potentially be stimulated through multiple upstream mechanisms.
Similarly, the older males demonstrated significantly higher LC3-II with treatments of bafilomycin with rapamycin at the 37.0 and 36.5 °C conditions (Figure 4B).However, significantly higher p62 in older compared to young males (Figure 4F) indicates that older adults are still displaying dysfunctional autophagic activity, which is also supported by an increase in apoptotic signaling (Figure 5B).Despite PBMCs from older males signaling an increase in apoptosis from an influx in cleaved-caspase-3 during all in vitro cold conditions (Figure 5B), the increase in LC3-II during the 36.5 °C of in vitro cooling is a sign of slight improvement in autophagic signaling as it contrasts the absence of any change in LC3-II during in vivo cooling in either condition (Figure 1B).This may suggest that the single dose of rapamycin was insufficient to surpass the requirements of autophagic-induced apoptotic suppression [35] during cold exposure.Others have utilized repeated dosages of autophagic stimulation (i.e., via rapamycin or trehalose treatment) as well as long-duration fasting to initiate autophagy which improved cold tolerance.For example in zebrafish, 48 h of fasting (to initiate starvation-induced autophagy) increased survival rates at 36 and 48 h of 11 °C cold-water exposure, however, it was not until 96 h of fasting where survival rates increased at 60 h of 11 °C cold-water exposure. [17]Further in mesenteric arteries of mice with type II diabetes (a disease associated with reduced autophagy), [36] repeated treatments of rapamycin and trehalose (an autophagy inducer through increased transcription factor EB) over two weeks normalized LC3-II, beclin-1, and p62 to levels comparable to those of non-diabetic mice. [37]Although this investigation assessed a single dose of rapamycin treatment, it would be of great interest to determine whether repeated exposures to rapamycin (or other strategies such as repeated cold exposures via an acclimation) would similarly initiate autophagic activity during cold exposure in older males to potentially restore cold-induced autophagy.
Considering both in vivo and in vitro cold exposure expressed increased simultaneous HSP90 and autophagic activity in young but not older males (Table 1), the presence of HSP90 may be required under certain stress conditions to activate autophagy.For example, inhibition of HSP90 in several lung cancer cell lines (i.e., NSCLC) showed to impede autophagic activity by reducing the expression of ATG7 (Autophagy Related Gene 7). [38]In the present investigation, the stimulation of autophagy through rapamycin may work in a feedback loop with the HSR, although the mechanisms responsible for the HSP90 and autophagy rela-tionship would need to be explored in future studies.Further, although HSP90 only increased in young males, there were neither changes in IL-6 nor TNF- for either the young or older males (Table 2).The absence of increased intracellular inflammatory markers in the isolated environment of in vitro cooling may indicate that the associated inflammatory response observed in older adults in vivo is related to a systemic source (e.g., age-related lowgrade inflammation or the additive oxidative stress originating from working muscles during shivering [33,39] ).However, identifying the source of increased inflammation in older males during cold exposure requires further investigation.
There are a few considerations that should be noted.As a methodological design choice in the in vivo cold exposure protocol, clamping core temperature of each individual required the exposure water temperature and duration of immersion to vary marginally between conditions and between individuals.As a result, individuals were exposed to different absolute cold exposure temperatures, but were exposed to the same relative change in core temperature (and therefore level of thermal strain) for the respective condition (i.e., the maximum temperature exposure for PBMCs).Although there is a well-established difference in the peripheral vasoconstriction response to cold exposure, [3] it is unclear whether there are age-related differences in shivering patterns. [40,41]Additionally, there are also large inter-individual differences of unknown origin (although speculated to be due to genetic variation) in shivering frequencies and intensities in individuals [2] which may have a resulting impact on the cellular stress responses.Further, due to the tightly regulated process of autophagy and the high potential for dangerous and unintended consequences such as inhibiting global protein synthesis, nucleotide production, and lipid metabolism, [42] it is difficult to appropriately dose pharmacological agents which stimulate whole-body autophagy that are considered safe for human consumption. [43]As such, targeting the inhibition of mTOR to stimulate autophagy with rapamycin during natural stimuli such as cold exposure must be performed in vitro in humans.Additionally, although the inhibition of autophagy using bafilomycin may allow for higher resolution of protein quantification related to autophagy, this may also alter downstream and feedback responses within the PBMCs as seen with increased cleavedcaspase-3 in young males during the 36.5 °C condition with a simultaneous significant increase in LC3-II (Figures 4A and 5A).It should also be noted that the point or points in the autophagic process where age-related dysregulation is occurring during cold exposure cannot be determined with the present investigation.
In conclusion, autophagy responds in an intensity-dependent manner during moderate cold stress conditions, but higher intensity cold stress conditions inhibit autophagy activation and stimulate apoptotic pathways in young males.However, there exists a clear age-related decline of autophagy under cold stress conditions that promotes apoptotic cell death.Stimulating components of the autophagic machinery through rapamycin during cold exposure indicates that age-related autophagic dysfunction may be reversed, as evidenced by increases LC3-II levels observed in both young and older males.By demonstrating the relationship between autophagy, apoptosis, the HSR, and inflammation during cold exposure in young and older males, this investigation provides novel insights into the potential mechanisms that con-tribute to age-related intolerance to cold exposure and the mechanistic potential to reverse the related autophagic dysfunction.

Experimental Section
All study procedures were approved by the University of Ottawa Health Sciences and Science Research Ethics Board (H01-20-5404) and agreed with the latest version of the Declaration of Helsinki.Written informed consent was obtained by all study volunteers prior to participation.
Human Participants: Forty habitually active (physically active at least 3 days per week for a minimum of 30 min per session) males who were free of known illnesses or disease participated.Of the 40 participants, 10 young (mean ± SD; age: 24 ± 3 years, maximal oxygen consumption (VO 2max ): 47.8 ± 4.9 ml kg min −1 , body surface area: 1.44 ± 0.19 m 2 ) and 10 older (age: 61 ± 3 years, VO 2max : 36.2 ± 5.5 ml kg min −1 , body surface area: 1.51 ± 0.17 m 2 ) males underwent two bouts of cold-water immersion, performed on separate days (described below).The remaining 20 participants were age-matched (± 1 year) to the cold-water immersion groups (24 ± 3 years and 61 ± 3 years for young and older males, respectively) and participated in the cell experiments (described below).All participants were instructed to refrain from physical activity, alcohol, caffeine, and antiinflammatory drugs for up to 8 h prior to their participation.Further, participants were instructed to arrive at the laboratory in a post-absorptive state by consuming a light-to-moderate meal (≈400-500 kcal) 2 h prior to their arrival.Participants were also instructed to consume a calorically similar meal for each experimental session.
Cold Water Immersion: Each participant underwent one preliminary session and two experimental sessions, separated by at least 72 h.During the preliminary session, participants' height and body mass were measured.Body surface area was calculated using the equation by Du Bois & Du Bois. [44]Participants then performed an incremental cycling maximal exercise test [45] to determine their VO 2max (MCD Medgraphics Ultima Series, MGC Diagnostics, MN, USA).
Two experimental sessions were performed in a randomized order on separate days at least 72 h apart.Upon arrival to the laboratory for each experimental session, participants changed into swim shorts and were instrumented with an esophageal core temperature probe by inserting a general-purpose thermocouple temperature probe (Mon-a-therm General Purpose Temperature Probe; Mallinckrodt Medical, St. Louis, MO, USA), positioned at the level of the heart.Participants remained seated in a thermoneutral environment (≈23 °C) for ≈30 min to allow for stabilization of core temperature.Following the 30 min habituation period, venous blood samples were collected in EDTA vacutainer tubes (BD Vacutainer; Becton Dickinson, Franklin Lakes, NJ) via antecubital venipuncture.Participants were then immersed to the upper shoulders in circulating water (ranging from 12-20 °C) in a seated position to elicit a moderate cold stress characterized by a decrease in core temperature by 0.5 °C (moderate cold stress) from baseline measurements or a high cold stress characterized by a decrease in core temperature by 1.0 °C (high cold stress) from baseline resting measurements. [2]A relative change in core temperature was chosen to define cold stress intensities to account for day-to-day variation in core temperature that may curtail or exaggerate the physiological stress between conditions.Further, unlike heat stress, afferent signaling relating to cold stress responds to relative changes in temperature [46] which may alter the efferent physiological input (e.g., shivering), and therefore perceived cellular stress.The respective target temperature was maintained for 30 min by gradually raising the water temperature to ≈20-25 °C before a secondary blood sample was obtained.Thereafter, the circulating water was immediately warmed to 39°C for rapid rewarming until the participants' core temperature returned to baseline values.The participants were then removed from the tank of water, towel dried, and changed into dry clothing before consuming a self-provided meal (≈400-600 kcals) 1 h following the cessation of cooling (participants were instructed to bring the same meal for both experimental sessions).They then remained in a thermoneutral room (≈23 °C) until a final blood sample was obtained, 3 h after the rewarming began.
Peripheral Blood Mononuclear Cell Isolation: Immediately following each blood draw, whole blood was suspended on histopaque (Histopaque-1077, Sigma-Aldrich) and centrifuged at 800 relative centrifugal force (RCF) for 30 min at 22 °C.The PBMCs were isolated and suspended in phosphate-buffered saline (PBS, P4417, Sigma-Aldrich) before centrifuging at 1100 RCF for 10 min at 4 °C.The supernatant was aspirated and the PBMC pellet was resuspended in PBS for a final centrifuging at 12 500 RCF for 10 min at 4 °C.The supernatant was aspirated prior to storage at −80 °C until analysis.
Statistical Analysis: Protein quantification results are expressed as means ± SEM.Physiological data (including esophageal and water temperatures, time to reach target temperature, and duration of the rewarming period) is expressed as means ± SD.To assess descriptive changes in water temperature and core temperature, data were each analyzed using a one-way ANOVA and two-tailed t-tests.To assess the ef-fect of cold-water immersion on cellular function, protein quantifications of each protein were analyzed with linear mixed-models to evaluate the effects of experimental phase (baseline, end of cold exposure, and recovery) and condition (moderate and high cold stress).To assess the effect of in vitro cold exposure on cellular function, protein quantifications were analyzed with linear mixed models to evaluate the effects of treatment (DMSO control, bafilomycin, rapamycin, and bafilomycin with rapamycin) and temperature condition (37.0, 36.5, 36.0, and 35.0 °C).Similarly, to assess the effect of aging, core temperature and protein quantifications were analyzed using linear mixed-models to evaluate the effects of experimental phase (baseline, end of cold exposure, and recovery) or treatment (DMSO, bafilomycin, rapamycin, and bafilomycin with rapamycin) and age (young and older) with independent analyses performed for each condition (moderate cold stress, high cold stress, 37.0, 36.5, 36.0, and 35.0 °C).Tukey's post-hoc comparisons were performed in the event of a significant interaction.All statistical analyses were performed using Prism (GraphPad Prism, version 8.4) with an alpha level set at 0.05 to indicate statistical significance.

Figure 1 .
Figure 1.The change in core (esophageal) temperature (ΔT eso ; panels A & B) and absolute T eso (panels C & D) during the 30 min clamping period at the end of the moderate cold stress (a 0.5 °C reduction in core temperature below resting levels) and high cold stress (a 1.0 °C reduction in core temperature below resting levels) cold exposure conditions in young A and C) and older B and D) males.Data are presented as means ± SD (n = 10 young and 10 older males).*Indicates significantly higher core temperature in the moderate than high cold stress condition (p < 0.05).

Figure 2 .
Figure 2. Relative quantity (RQ) of protein content of LC3-II (panels A & B), LC3-II/LC3-I (panels C and D), p62 (panels E and F), and beclin-2 (G and H) in young (left) and older (right) males immediately post and 3-hours (3 h) following cold-water immersion to elicit a decrease in core temperature by 0.5 °C (moderate cold stress) or 1.0 °C (high cold stress) below resting levels.Representative proteins analyzed through Western blotting are presented in panel I. Data are presented as means ± SEM (n = 10 young and 10 older males) with the main effect and interaction p-values presented in the corner of their respective proteins.*Indicates significantly different than baseline (p < 0.05).# indicates significantly different in the older compared to the young males (p < 0.05).

Figure 3 .
Figure 3. Relative quantity (RQ) of protein content of cleaved-caspase-3 (panels A and B) and cleaved/pro-caspase-3 (panels C and D) in young (left) and older (right) males immediately post and 3-hours (3 h) following cold-water immersion to elicit a decrease in core temperature by 0.5 °C (moderate cold stress) or 1.0 °C (high cold stress) below resting levels.Representative proteins analyzed through Western blotting are presented in panel E. Data are presented as means ± SEM (n = 10 young and 10 older males) with the main effect and interaction p-values presented in the corner of their respective proteins.*Indicates significantly higher than baseline (p < 0.05).# indicates significantly higher in the older compared to the young males (p < 0.05).

Figure 4 .
Figure 4. Relative quantity (RQ) of protein content of LC3-II (panels A and B), LC3-II/LC3-I (panels C & D), and p62 (panels E and F) in peripheral blood mononuclear cells isolated from young (left) and older (right) males immediately following 2 h of in vitro cold exposure at 37.0, 36.5, 36.0, and 35.0 °C with treatments of DMSO control (BL), bafilomycin (BAF), rapamycin (RAPA), and bafilomycin with rapamycin (BAF+RAPA; B+R).Representative proteins analyzed through Western blotting are presented in panel G. Data are presented as means ± SEM (n = 10 young and 10 older males) with the main effect and interaction p-values presented in the corner of their respective proteins.*Indicates significantly different than baseline (p < 0.05). Indicates significantly higher bafilomycin and rapamycin than bafilomycin of the same temperature condition.# indicates significantly different in the older compared to the young males (p < 0.05).

Figure 5 .
Figure 5. Relative quantity (RQ) of protein content of cleaved-caspase-3 (panels A and B) and cleaved/pro-caspase-3 (panels C and D) in peripheral blood mononuclear cells isolated from young (left) and older (right) males immediately following 2 h of in vitro cold exposure at 37.0, 36.5, 36.0, and 35.0 °C with treatments of DMSO control (BL), bafilomycin (BAF), rapamycin (RAPA), and bafilomycin with rapamycin (BAF+RAPA; B+R).Representative proteins analyzed through Western blotting are presented in panel E. Data are presented as means ± SEM (n = 10 young and 10 older males) with the main effect and interaction p-values presented in the corner of their respective proteins.*Indicates significantly higher than baseline (p < 0.05).# indicates significantly higher in the older compared to the young males (p < 0.05).

Table 1 .
Changes in protein content from baseline during in vivo cold exposure with representative protein blots.

Table 2 .
Changes in protein content from baseline during in vitro cold exposure with representative protein blots.