Continued clearance of apoptotic cells critically depends on the phagocyte Ucp2 protein

doi:10.1038/nature10340

Nature. Author manuscript; available in PMC 2012 December 4.

Published in final edited form as:

Nature. 2011 August 21; 477(7363): 220–224.

doi: 10.1038/nature10340

PMCID: PMC3513690

NIHMSID: NIHMS422520

Continued clearance of apoptotic cells critically depends on the phagocyte Ucp2 protein

Daeho Park,^1,² Claudia Han,^1,^2,³ Michael R. Elliott,^1,² Jason M. Kinchen,^1,^2,³ Paul C. Trampont,^1,² Soumita Das,⁴ Sheila Collins,⁵ Jeffrey J. Lysiak,⁶ Kyle L. Hoehn,⁷ and Kodi S. Ravichandran^1,^2,³

¹Center for Cell Clearance, University of Virginia, Charlottesville, Virginia 22908, USA

²Beirne B. Cater Center for Immunology Research, University of Virginia, Charlottesville, Virginia 22908, USA

³Department of Microbiology, University of Virginia, Charlottesville, Virginia 22908, USA

⁴Department of Medicine, University of Virginia, Charlottesville, Virginia 22908, USA

⁵Diabetes and Obesity Research Center, Sanford-Burnham Medical Research Institute, Orlando, Florida 32827, USA

⁶Department of Urology, University of Virginia, Charlottesville, Virginia 22908, USA

⁷Department of Pharmacology, University of Virginia, Charlottesville, Virginia 22908, USA

The publisher's final edited version of this article is available at Nature

See other articles in PMC that cite the published article.

Abstract

Rapid and efficient removal of apoptotic cells by phagocytes plays a key role during development, tissue homeostasis, and in controlling immune responses^1–5. An important feature of efficient clearance is the capacity of a single phagocyte to ingest multiple apoptotic cells successively, and to process the increased load of corpse-derived cellular material^6–9. However, factors that influence sustained phagocytic capacity or how they in turn influence continued clearance by phagocytes are not known. Here we identify that the ability of a phagocyte to control its mitochondrial membrane potential is a critical factor in the capacity of a phagocyte to engulf apoptotic cells. Changing the phagocyte mitochondrial membrane potential (genetically or pharmacologically) significantly affected phagocytosis, with lower potential enhancing engulfment and higher membrane potential inhibiting uptake. We then identified that Ucp2, a mitochondrial membrane protein that acts to lower the mitochondrial membrane potential^10–12, is upregulated in phagocytes engulfing apoptotic cells (but not synthetic targets, bacteria, or yeast). Loss of Ucp2 limited the capacity of phagocytes to continually ingest apoptotic cells, while overexpression of Ucp2 increased the capacity for engulfment and the ability to engulf multiple apoptotic cells. Mutational and pharmacological inhibition of Ucp2 uncoupling activity reversed the positive effect of Ucp2 on engulfment capacity, suggesting a direct role for Ucp2-mediated mitochondrial function in phagocytosis. Macrophages from Ucp2-deficient mice^{13, 14} were impaired in their capacity to engulf apoptotic cells in vitro, and Ucp2-deficient mice displayed profound in vivo defects in clearing dying cells in the thymus and the testes. Collectively, these data suggest that phagocytes alter the mitochondrial membrane potential during engulfment to regulate uptake of sequential apoptotic cells, and that Ucp2 is a key molecular determinant of this step in vivo. Since Ucp2 function has also been linked to metabolic diseases and atherosclerosis^14–16, these data identifying a new role for Ucp2 in regulating apoptotic cell clearance may provide additional insights toward understanding the complex etiology and pathogenesis of these diseases.

When one cell engulfs another, the phagocyte essentially doubles its cellular contents. Remarkably, phagocytes often seem to be capable of ingesting and processing multiple targets sequentially. However, the cellular features that regulate the ‘capacity’ of a given phagocyte to eat apoptotic cells, or regulate sequential phagocytosis of multiple dying cells are not understood. Since metabolites derived from the ingested apoptotic cells are likely to influence mitochondrial function in the phagocyte, we first asked whether there are any alterations to the mitochondrial membrane potential within phagocytes that have ingested apoptotic cells. Phagocytes labeled with the dye JC-1 (whose ratio of red to green fluorescence is dependent on mitochondrial membrane potential¹⁷ and reversible, Supplementary Fig 1a), or another dye TMRE were incubated with apoptotic thymocytes in a time course. To focus specifically on phagocytes with internalized targets, we labeled the apoptotic thymocytes with a pH-dependent fluorescent dye, cypHer5E. The cypHer5E fluorescence of the labeled cells is significantly enhanced during phagocytosis, after the apoptotic corpse enters the acidic environment of the phagolysosome¹⁸. The relative mitochondrial membrane potential within phagocytes that contained engulfed apoptotic cells was significantly increased (1.99-fold ±0.11, p < 0.001, n=6) compared to those phagocytes without ingested targets within the same population (Fig. 1a and Supplementary Fig. 1b). It is noteworthy that the contribution to the enhanced mitochondrial membrane potential signal seen within phagocytes from the mitochondria of ingested targets (i.e. apoptotic thymocytes) was minimal (Supplementary Fig. 1c). Interestingly, there was no increase in the mitochondrial membrane potential within phagocytes that had ingested synthetic targets (2 µm carboxylate-modified beads) (Fig. 1a and Supplementary Fig. 1b). Since these synthetic targets have previously been shown to use components of the apoptotic cell engulfment machinery¹⁹, yet do not bring in a metabolic load for the phagocytes, these data suggest that the increase in relative mitochondrial membrane potential is specific to the uptake of cellular material from apoptotic cells, and not a general effect of engulfment.

Figure 1

Ucp2 is upregulated in phagocytes during engulfment and affects uptake of apoptotic cells

We then determined the duration of this increased mitochondrial membrane potential within a phagocyte after engulfment of apoptotic cells. After the initial incubation of apoptotic thymocytes with the phagocytes for 3 hours (to allow time for engulfment), the unbound or excess targets were washed off, and we monitored the mitochondrial membrane potential in a time course. Following the engulfment phase, mitochondrial membrane potential decreased over time, returning to the baseline level of non-engulfing phagocytes by 2 hours during this chase time (Fig. 1b). Thus, there is a transient increase in the mitochondrial membrane potential within the phagocytes with ingested apoptotic cells.

In general, cellular ATP levels are maintained within a fairly narrow range and we found this to be true in engulfing phagocytes despite their increased metabolic load. Engulfing phagocytes had a modest 10 % increase in cellular ATP levels only at the 2 hour time point after engulfment (Fig. 1c). The lack of ATP generation concomitant with a decrease in mitochondrial membrane potential subsequent to engulfing apoptotic cells suggested that phagocytes possess a rapid mechanism for uncoupling nutrient oxidation from ATP generation¹¹.

A well-known endogenous mechanism for uncoupling the proton gradient and ATP generation is mediated by uncoupling proteins (UCPs)^{11, 20}. The UCPs regulate the mitochondrial membrane potential via dissipation of the proton gradient across the inner mitochondrial membrane, without generation of ATP. Although the function of Ucp1 in mitochondrial uncoupling and heat generation is well described²¹, Ucp1 expression is restricted to brown adipose tissue. Since Ucp2 is more ubiquitously expressed¹¹, including in multiple key primary phagocytic cell populations (Supplementary Fig. 2a), we examined its possible involvement in phagocytes during the engulfment of apoptotic cells. Interestingly, while the mRNA level of Ucp2 was not detectably altered, the mitochondrial Ucp2 protein levels in phagocytes were increased during engulfment, in a time course after incubation with apoptotic cells (Fig. 1d left and Supplementary Fig. 2b–c). It is important to note that the increased Ucp2 levels were not derived from the targets being ingested as the levels of Ucp2 in thymocytes were below detection and feeding Ucp2 deficient apoptotic thymocytes to the phagocytes also resulted in enhanced Ucp2 levels in phagocytes (Fig. 1d and Supplementary Fig. 2d). Under the same conditions, Ucp2 protein levels did not change in phagocytes engulfing synthetic targets, which is consistent with a lack of effect of synthetic targets on mitochondrial membrane potential (Fig. 1d right).

To directly address whether Ucp2 may regulate engulfment of apoptotic cells, we knocked down endogenous Ucp2 expression in phagocytes by siRNA (Fig. 1e). We first noted that the decreased levels of Ucp2 within NIH/3T3 cells led to increased mitochondrial membrane potential both at the basal state and during engulfment (Fig. 1f and Supplementary Fig. 2e). We also observed two key phenotypes in engulfment by Ucp2-depleted cells. First, the Ucp2 knockdown phagocytes were less efficient engulfers, as measured by the percentage of phagocytes ingesting apoptotic cells over a 6-hour time course (Fig. 1g left). Second, the Ucp2 knockdown cells showed reduced phagocytic ‘capacity’, i.e. lower amount of apoptotic material ingested per phagocyte, as measured by the mean fluorescence intensity (MFI) derived from the labeled apoptotic cells (Fig. 1g right). It is noteworthy that the phagocytes with Ucp2 knockdown showed no defect in engulfing synthetic targets and the number of particles taken up per phagocyte was also unaltered (Fig. 1h). Collectively, these data suggest that phagocytes engulfing apoptotic cells upregulate the Ucp2 protein and that Ucp2 expression level contributes to the capacity of a phagocyte to ingest apoptotic cells.

The above data prompted us to ask whether Ucp2 overexpression in phagocytes may provide a ‘gain of function’ for engulfing apoptotic cells. We transiently expressed Flag-tagged Ucp2 in LR73 phagocytes. Ucp2^Flag was detected in the mitochondrial fraction (and not in the phagosomes), and the overexpressed Ucp2^Flag co-localized with another mitochondrial inner membrane protein, AIF, indicating proper targeting of ectopically expressed Ucp2 (Fig. 2a and Supplementary Fig. 3). Ucp2 overexpressing phagocytes showed three striking phenotypes with respect to engulfment. First, a greater percentage of the Ucp2-transfected LR73 cells ingested apoptotic cells compared to the control transfected cells; this enhanced uptake was specific for apoptotic cells and not seen with synthetic targets, and depended on the known engulfment machinery components (Fig. 2b and Supplementary Fig. 4a–b). Second, in a time course engulfment assay, the Ucp2 overexpressing cells continued to ingest apoptotic targets compared to control transfected cells (Fig. 2c left). Moreover, while the control-transfected LR73 cells displayed a plateau in the accumulation of apoptotic cell-derived fluorescence beyond 2 hours, the Ucp2 transfected LR73 cells continued to accumulate more apoptotic cells up to 6 hours (as determined by the mean fluorescence intensity) (Fig. 2c right). This was also confirmed by microscopy, where Ucp2 overexpressing cells consistently displayed more ingested apoptotic cells per phagocyte compared to control transfected cells (Fig. 2d). Quantitation of the number of targets per phagocyte in this assay revealed that a higher fraction of Ucp2 expressing cells showed more than two apoptotic cells per phagocyte compared to GFP expressing cells (Fig. 2e). Third, there was no difference in the rate of uptake of synthetic targets over a time course between Ucp2 and control-transfected cells (Fig. 2f), further supporting the notion that Ucp2 plays a specific role in sensing the metabolic load derived from apoptotic cells. The increased uptake of apoptotic cells was also observed with overexpression of other homologues of Ucp2 (Supplementary Fig. 4c). Taken together, these data revealed that overexpression of Ucp2 provides a ‘gain of function’ for phagocytes and that Ucp2 levels can confer a capacity to ‘continue to eat’ apoptotic cells on phagocytes, which is dependant on the canonical engulfment machinery.

Figure 2

Ucp2 overexpression promotes continued uptake of apoptotic cells

We then asked whether the increased phagocytic capacity of Ucp2 overexpressing cells correlates with reduced mitochondrial membrane potential. As expected, Ucp2 cells showed decreased mitochondrial membrane potential compared to controls (assessed by JC-1 fluorescence) (Fig. 2g). In a second approach, we directly visualized the mitochondrial uptake of MitoTracker, a dye that accumulates as the mitochondrial membrane potential increases²². The MitoTracker signal was distinctly weaker in Ucp2^Flag expressing cells compared to neighboring non-transfected cells (Fig. 2h). These data indicate that Ucp2 overexpression decreases the mitochondrial membrane potential. In previous studies, the drug genipin has been shown to inhibit the proton leak or mitochondrial uncoupling mediated by Ucp2²³. Genipin strongly inhibited the enhanced engulfment seen in Ucp2-overexpressing cells (both the percentage and the capacity for engulfment) (Fig. 2i). As another approach, we generated a D28N mutant of Ucp2 that has previously been shown to decrease the uncoupling activity of Ucp2^{24, 25}. Although Ucp2^D28N was expressed comparably to wild-type Ucp2 and localized to the mitochondria, Ucp2^D28N failed to enhance the percentage of phagocytes engulfing apoptotic cells (Fig. 2j). These data suggest that a decrease in mitochondrial membrane potential by Ucp2 correlates with the ‘gain of function’ phenotype, manifested as enhanced and continued uptake of apoptotic cells by Ucp2 expressing phagocytes.

Since UCP2 is also reported to have other functions²⁶, we next asked whether the effect of Ucp2 could be mimicked by synthetic uncouplers that lower the mitochondrial membrane potential^{27, 28}. We initially tested two different synthetic uncouplers, 2,4-DNP and FCCP, over a range of concentrations where the total cellular ATP levels were unaltered and the phagocytes appeared morphologically similar to vehicle control treated cells (Supplementary Fig. 5a–b). Both 2,4-DNP and FCCP significantly enhanced the uptake of apoptotic thymocytes by LR73 phagocytes in a dose-dependent manner (Fig. 3a); this increased uptake correlated with the decreased mitochondrial membrane potential induced by these drugs (Supplementary Fig. 5c–d). To determine if mitochondrial membrane potential itself, i.e. independent of increased proton leak and increased electron transport chain flux, was the signal for enhanced engulfment capacity we tested sodium azide, an inhibitor of complex IV of the electron transport chain that slows electron transport flux and lowers mitochondrial membrane potential. Increasing concentrations of sodium azide that did not affect cellular ATP levels enhanced engulfment by phagocytes along with a progressive decrease in the mitochondrial membrane potential within the phagocytes (Fig. 3b and Supplementary Fig. 5e–f). Higher concentrations of sodium azide, which decrease the cellular ATP levels, did inhibit apoptotic cell engulfment (since engulfment is an ATP-dependent process) (Fig. 3b and Supplementary Fig. 5f). Importantly, synthetic uncoupling also enhanced the ability of phagocytes to continue to ingest apoptotic cells in a time course, similar to Ucp2 overexpression (Fig. 3c). This was also confirmed by microscopy and flow cytometry, which revealed that phagocytes treated with the synthetic uncoupler FCCP have an increased capacity to ingest multiple apoptotic cells (Fig. 3d–e).

Figure 3

Ucp2 links mitochondrial membrane potential to engulfment of apoptotic cells

We next asked whether artificially increasing the mitochondrial membrane potential of phagocytes would inhibit engulfment. Treatment of cells with oligomycin, an inhibitor of ATP synthase, increased mitochondrial membrane potential and potently inhibited engulfment (Fig. 3f). Notably, the concentration of oligomycin chosen for these experiments affected the mitochondrial membrane potential but not the total cellular ATP levels (Supplementary Fig. 5gh). In addition to LR73 cells, we also assessed the effect of the synthetic uncouplers DNP and FCCP, as well as sodium azide and oligomycin, on phagocytosis by NIH/3T3 cells; again, decreasing the mitochondrial membrane potential strongly promoted the uptake of apoptotic cells (data not shown).

We also asked whether modulating the metabolic state of the phagocytes would affect the engulfment of apoptotic cells. Intriguingly, culturing the phagocytes in glucose free medium, which decreased the mitochondrial membrane potential, resulted in enhanced engulfment of apoptotic cells (Fig. 3g and Supplementary Fig. 6a). Conversely, phagocytes cultured in the presence of excess of glucose (which showed higher mitochondrial membrane potential) blocked the enhanced uptake of apoptotic cells either by Ucp2 overexpression or by FCCP (Fig. 3h–i and Supplementary Fig. 6b–c). Taken together, these data suggest that the mitochondrial membrane potential within phagocytes is a critical determinant in regulating the phagocytic capacity and continued uptake by phagocytes.

Increased metabolic load in phagocytes during engulfment may affect many aspects of metabolism and mitochondrial signaling such as β-oxidation and reactive oxygen species (ROS) production. Since Ucp2 is known to increase β-oxidation and decrease ROS production¹¹ one hypothesis could be that excess cellular lipids or ROS feedback could slow down the engulfment pathway. First, we tested whether fatty acid oxidation may be altered during engulfment and how this relates to the phenotypes seen with Ucp2 over/under expression and changes to mitochondrial membrane potential. Incubation of phagocytes with apoptotic thymocytes increased the rate of fatty acid oxidation and this was not observed when phagocytes were fed synthetic targets (Fig. 3j–k). However, we did not find any correlation between β-oxidation and Ucp2 levels; cells overexpressing Ucp2 as well as Ucp2 deficient phagocytes displayed higher fatty acid oxidation rate (Supplementary Fig. 7a–b), even though they have opposite phenotypes with respect to their ability to engulf apoptotic cells. This suggested that lipid oxidation alone is not a major regulator of apoptotic cell clearance. Ucp2 has also been shown to negatively regulate ROS levels and decreasing ROS could also be a potential mechanism by which Ucp2 promotes continued engulfment. However, we could not establish a link between mitochondrial ROS levels in phagocytes and engulfment of apoptotic cells. Increasing phagocyte mitochondrial ROS levels via addition of the drugs rotenone or antimycin A (which block complexes I or III within the electron transport chain, respectively) did not decrease apoptotic cell engulfment (Supplementary Fig. 8a–b). In fact, these drugs modestly increased the engulfment of apoptotic cells, likely due to decreased mitochondrial membrane potential. Furthermore, neither ameliorating ROS with FCCP, nor scavenging ROS with well-known scavengers Tiron or MitoTEMPO enhanced the ability of phagocytes to engulf apoptotic cells (Supplementary Fig. 8a–f). Furthermore, overexpression of the mitochondrial antioxidant enzyme superoxide dismutase 2 (SOD2) did not affect engulfment of apoptotic cells (Supplementary Fig. 8g). We also noted that the master regulatory transcription factor for mitochondrial biogenesis, PGC1α, did not change during engulfment and overexpression of PGC1α in phagocytes did not promote engulfment of apoptotic cells (Data not shown and Supplementary Fig. 8h). Moreover, neither the AMPK nor the mTOR signaling pathways were activated during engulfment, which was determined by phosphorylation of AMPK and P70S6K, respectively (Supplementary Fig. 8i–j). Collectively, while these data cannot completely rule out some contribution of β-oxidation and ROS levels in regulating apoptotic cell clearance, none of these pathways in isolation could account for the engulfment phenotype regulated by Ucp2. Rather, the Ucp2 mediated regulation of the phagocyte mitochondrial membrane potential correlated best with the ability of phagocytes to continue to engulf apoptotic cells.

To determine whether the recognition and uptake of apoptotic cells via specific engulfment receptors may be sensed or integrated into Ucp2/mitochondrial signaling, we generated LR73 cells overexpressing the phosphatidylserine receptor Tim-4. Consistent with previous reports^{8, 29, 30}, overexpression of the engulfment receptor Tim-4 led to increased uptake of apoptotic cells. However, we found three pieces of data that suggested a link between Tim-4 mediated apoptotic cell recognition and Ucp2. First, while there was no difference in the basal Ucp2 level between control and LR73 cells overexpressing Tim-4, when incubated with apoptotic cells, the Tim-4 overexpressing cells upregulated Ucp2 to a much higher level (Fig. 3l top). Importantly, incubation with synthetic targets did not increase Ucp2 expression in phagocytes, even though the uptake of these targets is also promoted by Tim-4 overexpression (Fig. 3l bottom). Second, the Tim-4 overexpressing cells also continued to take up multiple apoptotic cells (indicated by MFI). This continued uptake, without reaching a plateau that is seen in the control cells, is consistent with the increased upregulation of Ucp2 in Tim-4 overexpressing cells and very similar to LR73 cells transfected with Ucp2 (Fig. 3m). Third, an Ucp2 inhibitor drug genipin blocked the increased uptake of apoptotic cells seen due to Tim-4 (Fig. 3n). This prompted us to ask whether the enhanced engulfment seen with synthetic uncoupling or Ucp2 overexpression could be due to enhanced levels of Tim-4 surface expression. However, the surface expression of endogenous Tim-4 or HA-Tim-4 was not affected by Ucp2 overexpression, Ucp2 deficiency, synthetic uncoupling or by scavenging ROS (Supplementary Fig. 9a–e). Collectively, these data suggest a link between apoptotic cell recognition at the phagocyte membrane and the mitochondrial membrane potential with Ucp2 serving as a key molecular intermediate.

We next addressed the relative importance of Ucp2 in regulating the mitochondrial membrane potential and apoptotic cell clearance in vivo using Ucp2 deficient mice^{13, 16}. Bone marrow-derived macrophages (BMDMs) from Ucp2^−/− mice had a higher relative mitochondrial membrane potential compared to BMDMs from Ucp2^+/+ littermates (Fig. 4a). We then assessed the in vitro phagocytic capacity of BMDMs from Ucp2^+/+ or Ucp2^−/− mice to engulf different targets and found that in a time course of engulfment, Ucp2^−/− macrophages were consistently less efficient and showed lower phagocytic capacity (Fig. 4b). By contrast, the BMDMs from Ucp2^−/− mice showed no defect in the uptake of synthetic targets, either live or dead bacteria or zymosan A particles (Fig. 4c and Supplementary Fig. 10a–f). This suggested that during apoptotic cell clearance Ucp2 requires both the metabolic load (absent in synthetic targets) and sensing of the nature of entry of targets (e.g. bacteria and yeast, which do carry a metabolic load but are taken up via other types of receptors). It is notable that there was still some level of basal engulfment that was unaffected in the Ucp2 null phagocytes, which may in part be explained by the upregulation of Ucp3 that was observed in 6-day cultures of both splenocytes and BMDMs from Ucp2^−/− mice (Supplementary Fig. 11a–c).

Figure 4

Impaired phagocytosis and clearance of apoptotic cells in Ucp2 deficient mice

We then asked how Ucp2^−/− phagocytes would clear apoptotic cells in vivo under conditions where a significant population of cells within a tissue is undergoing apoptosis. Injection of dexamethasone (Dex) in mice induces rapid and synchronous death of thymocytes, and the subsequent clearance of apoptotic thymocytes by resident phagocytes can provide a reproducible and quantitative in vivo model of apoptotic cell clearance³¹. Using this model, we tested the efficiency of apoptotic cell clearance in control and Ucp2^−/− mice. Compared to control mice, which show a decrease in overall thymic size at 6 hours after Dex injection, the Ucp2^−/− mice showed only a limited reduction in thymic size after Dex injection (Fig. 4d). Assessing the absolute numbers of cells in the thymi confirmed that the Dex-injected Ucp2^−/− mice had much higher total thymic cell number than that of Dex-injected Ucp2^+/+ mice (Fig. 4e). Importantly, we found that thymocytes from Ucp2^−/− mice underwent apoptosis to the same extent as the control littermates ex vivo (Supplementary Fig. 12a–b); moreover, the migration of monocytes or macrophages toward find-me signals from apoptotic cells was not affected by Ucp2 levels or mitochondrial uncoupling (Supplementary Fig. 13a–e). These data suggests that increased thymic cellularity most likely resulted from defective clearance in the Ucp2^−/− mice. We then determined the status of uncleared apoptotic cells in the thymus of Ucp2^+/+ and Ucp2^−/− mice by TUNEL, and found a consistently higher level of unengulfed apoptotic cells within the thymi of Dex treated Ucp2^−/− mice without altered density of F4/80 positive cells (Fig. 4f and Supplementary Fig. 13f–h). Quantitation of the TUNEL fluorescence over multiple thymic sections of several mice revealed a three-fold increase in the amount of TUNEL staining in the thymi of Dex-treated Ucp2^−/− mice compared to Ucp2^+/+ controls (Fig. 4g). We also tested the ability of Ucp2^−/− mice to clear apoptotic cells in other tissues in vivo; acute apoptosis of germ cells in the testes was induced by testicular torsion and the clearance of these dying germ cells by the Sertoli cells of the testes was assessed. The testes of Ucp2^−/− mice contained the increased number of uncleared apoptotic cells per seminiferous tubule compared to Ucp2^+/+ (Fig. 4h–i). These data demonstrate an essential role for Ucp2 in apoptotic cell clearance in vivo.

While significant progress has been made in recent years regarding how phagocytes recognize and engulf apoptotic cells^32–34, the data presented in this report provide several new insights toward our understanding of the dynamic nature of apoptotic cell clearance. First, our studies using multiple approaches suggest that the mitochondrial membrane potential critically controls how well a phagocyte can engulf, with an inverse correlation between mitochondrial membrane potential and the capacity for engulfment. Second, these data identify the mitochondrial protein Ucp2 as a critical molecular determinant of the mitochondrial membrane potential within phagocytes during apoptotic cell engulfment. At the organism level, genetic ablation of Ucp2 in mice led to severe defects in the ability to clear apoptotic cells in the thymus and the testes. Thus, these data assign a new role to the mitochondria within phagocytes in controlling the engulfment capacity and identify Ucp2 as a key molecular rheostat of phagocyte engulfment capacity. Third, our data identify a previously unappreciated crosstalk between the mitochondria and the engulfment machinery within phagocytes. Our work establishes that ‘sensing’ of the total mitochondrial membrane potential critically influences the capacity of phagocytes to engulf apoptotic cells, and identify Ucp2 as a key determinant in phagocytes for continued uptake of apoptotic cells. This has broad implications for apoptotic cell clearance in vivo since failed clearance of apoptotic cells has been linked to inflammation and autoimmune diseases.

Since uncoupling proteins regulate the energetics within the mitochondria^{15, 35–37}, understanding the relationship between metabolic diseases and uncoupling proteins is an area of intense study^38–42. Loss of Ucp2 expression has been linked to impaired pancreatic β cell function and glucose-induced insulin secretion. The data presented here provide a new and unexpected link between mitochondrial function and the process of apoptotic cell clearance. Thus, our new observations that Ucp2 protein levels can be modulated downstream of engulfment receptors may be relevant toward understanding complex etiology of some of these metabolic diseases. Loss of Ucp2 has been linked to atherosclerosis, with mice deficient Ucp2 displaying a greater severity of disease^{14–16, 43}; similarly, failed clearance of apoptotic cells within atherosclerotic plaques and the resultant inflammatory milieu have been associated with the severity of atherosclerosis. Thus, our studies providing a link between Ucp2 and apoptotic cell clearance may help identify a possible mode of treatment for atherosclerosis and other diseases by enhancing the phagocytic potential via regulation of Ucp2 activity.

Methods

Cell culture and transfections

LR73 cells were cultured in Alpha-MEM, while NIH/3T3 cells and J774 macrophage cells were maintained in DMEM, and Jurkat cells were culture with RPMI, along with 10 % FBS and 1 % penicillin-streptomycin-glutamine. SCI cells were cultured in alpha-MEM and 1 mM of Na-pyruvate. LR73 cells were transfected using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. NIH/3T3 cells were nucleofected using amaxa nucleofector (kit R, program U-030) or transfected with Lipofectamine 2000 according to the manufacturer’s manual.

Primary cell culture

Bone marrow cells from 6-week old mice were culture in RPMI medium with 10 % FBS, 1 % penicillin-streptomycin-glutamine, and 10 % L929 cell conditioned medium and then plated. Two or six days later, more than 85 % of the adherent cells were F4/80 positive and 95 % of the cells were CD11b positive. Resident peritoneal cells were collected from 6-week-old mice and plated with RPMI supplemented with 10 % FBS and 1 % penicillin-streptomycin-glutamine. Four hours later, floating cells were washed with warm PBS twice and adherent cells were regarded as resident peritoneal macrophages and used in further assays.

Plasmids and mutagenesis

All constructs generated were sequenced to confirm fidelity and the presence of the appropriate mutations. The pEBB-Ucp2 and pEBB-Ucp2-Flag constructs were generated from the Ucp2 cDNA templates by a PCR-based strategy in the pEBB-Flag vector. pEBB-Ucp2^D28N was generated by site-directed mutagenesis. ELMO, Dock180, Rac, BAI1, and Tim-4 constructs used in this study have been described previously¹⁹.

Immunoblotting and antibodies

The antibodies used were anti-Ucp2 (Santa Cruz Biotechnology Inc., C-20), anti-Flag (Sigma, M-2), anti-Porin (Abcam, ab15895), anti-Erk2 (Santa Cruz, C-14),, and anti-AIF (Cell Signaling, D39D2). The phospho-AMPKα (Thr172), AMPKα Antibody, phospho-p70 S6 Kinase (Thr389), p70 S6 Kinase, were purchased from Cell Signaling Technology. To detect endogenous Ucp2, it is necessary to isolate mitochondria. For the isolation of mitochondria, the cells were resuspended in 500 µl of TS buffer (10 mM Tris, pH 7.5, 250 mM sucrose, 2 µg/ml DNase and protease inhibitor cocktail), and then subjected to three cycles of 5-minute freezing in liquid nitrogen followed by 10-minute thawing at 37 °C. Unbroken cells and nuclei were removed by centrifugation at 800g for 10 min and mitochondria were collected from the supernatant by centrifugation at 10, 000g for 20 min. The isolated mitochondria were lysed in RIPA buffer and subjected to immunoblotting. To measure induction of Ucp2 during apoptotic cell engulfment, 2.0 × 10⁶ LR73 cells were plated on the 100 mm culture dish two days before 1.0 × 10⁸ apoptotic thymocytes in 10 ml of alpha-MEM were added. After incubation for 3 hours, the mitochondria were isolated and the Ucp2 levels were analyzed. For BMDM, 5.0 × 10⁶ BMDM were plated on the 100 mm Petri-dish 1 day before the assay. Next day, 1.0 × 10⁸ apoptotic thymocytes in 10 ml of RPMI were added to BMDM and incubated for 1 hour. The cells were extensively washed with cold PBS and trypsinized and analyzed for Ucp2 levels. To detect overexpressed Ucp2, LR73 cells were transiently transfected with Flag tagged Ucp2 using Lipofectamine 2000 (Invitrogen). One day after transfection, the cells were lysed and subjected to immunoblotting against the respective tags or endogenous proteins.

Immunofluorescence microscopy

NIH/3T3 fibroblasts were plated on glass chamber slides and transfected with FLAG-Ucp2 either alone or in combination with YFP-Rab5 or YFP-Rab7. To stain lysosomes or mitochondria, cells were incubated with Lysotracker Red (1:10,000 dilution) or Mitotracker Deep Red (100 nM) in DMEM/10%FBS for 20 minutes. Cells were then fixed with 3% paraformaldehyde (Sigma) in PBS for 30 min, permeabilized with 0.1% Triton X-100 (Sigma) and blocked with 5% milk that had been clarified by high-speed centrifugation. Antibody staining was then performed using antibodies to FLAG (clone M5, Sigma) and/or AIF (D39D2, Cell Signaling) detected with either Alexa 488 or Alexa-555 goat anti-mouse and/or Alexa-555 goat anti-rabbit antibodies (highly cross-absorbed, Invitrogen), respectively. YFP was detected using an Alexa-488 anti-GFP antibody (Invitrogen). The stained cells were analyzed by Axio imager 2 with Apotome (Zeiss).

Quantitative RT-PCR

Total RNA was isolated using the RNeasy Mini Kit (Qiagen) and cDNA was generated from total RNA with Superscript III (invitrogen) according to the manufacturers’ protocol. Ucp1, Ucp2, and Ucp3 mRNA were detected by PCR using cDNA as template. For siRNA transfected NIH/3T3 cells and Ucp2 KO mice, relative Ucp2 or Ucp3 mRNA levels were determined by normalizing to Hprt using the StepOnePlus qPCR system (Applied Biosystems).

Phagocytosis assay

Phagocytosis assays were performed as described previously¹⁹. 1.0 × 10⁵ LR73 cells were transiently transfected in triplicates with the indicated plasmids either with GFP or GFP fusion proteins in a 24-well plate. The cells were incubated with apoptotic thymocytes or 2 µm carboxylate-modified red fluorescent beads (invitrogen), which mimic the negative charge on apoptotic cells and can serve as a simplified targets. For the induction of apoptosis, thymocytes were incubated with 50 µM Dexamethasone (Calbiochem Inc.) at 37 °C for 4 hours. The thymocytes were resuspended to a final concentration of 1.0 × 10⁷ cells/300 µl with Alpha-MEM (or cell culture medium for specific cell types) supplemented with 2% FBS and 0.2% penicillin-streptomycin-glutamine for the phagocytosis assay of LR73. The transfected cells were incubated with 300 µl of the apoptotic thymocyte resuspension in the 5% CO₂ incubator at 37 °C for desired times. After incubation of the phagocytes with targets, the phagocytes were extensively washed with cold PBS, trypsinized, resuspended in cold medium (with 1% NaN₃), and analyzed by two-color flow cytometry (FACSCalibur from BD). The transfected cells were recognized by their GFP fluorescence and targets were recognized by red fluorescence (carboxylate-modified red fluorescent beads or apoptotic thymocytes). Forward and side-scatter parameters were used to distinguish free unbound targets from phagocytes. The majority of ‘double-positive’ cells scored in the FACS assay represented targets engulfed by phagocytes, or targets in the process of being engulfed. The mean fluorescence intensity (MFI) in the red channel of the cells taking up targets provided an indication of the capacity of uptake (proportional to the number of particles taken up). For measuring phagocytosis by BMDM, 2.0 × 10⁴ bone marrow derived macrophages were plated in the 24-well plate 1 day before the phagocytosis assay. Next day, the cells were incubated with 2.0 × 10⁶ TAMRA-stained apoptotic thymocytes for desired times and analyzed for engulfment as above.

Internalization assay

Thymocytes were double stained with TAMRA (25 µM) and cypHer5E (1 µM, GE Healthcare) and phagocytosis assay was performed as described above. TAMRA and cypHer5E high double positive cells were considered as phagocytes internalizing the apoptotic cells, and TAMRA positive but cypHer5E low cells were regarded as bound but uninternalized targets on phagocytes.

Phagocytosis of dead Escherichia coli, staphylococcus aureus and Zymosan

2.0 × 10⁴ BMDMs from wild type or Ucp2 deficient mice were plated on a well of a 24-well plate. Next day, BMDMs were incubated with alex-488 conjugated 5 × 10⁶ E. coli and S. aureus or 5 × 10⁵ zymosan particles (invitrogen) for the desired times. Unbound targets were extensively washed with cold PBS five times and BMDMs were trypsinized and subjected to FACS analysis. Alex-488 positive BMDMs were considered as phagocytes engulfing targets.

Gentamicin protection assay for bacterial internalization

Quantitation of intracellular bacteria was done using the gentamicin protection assay. 2 × 10⁵ BMDMs/well were seeded into a 24-well culture dish 18 hours prior to infection at a multiplicity of infection of 10 for 1 hour in antibiotic-free media in a 37 °C CO₂ incubator. Cells were then washed and incubated with gentamicin for 90 min to kill extracellular bacteria. Subsequently, cells were lysed in 1% Triton-X 100, lysates were serially diluted, and plated directly onto Luria-Bertani agar plates. Total colony-forming units (CFUs) were enumerated the next day after overnight incubation at 37°C. Values were standardized to levels of colonization in control cell preparations.

Ucp2 knockdown

NIH/3T3 cells were nucleofected according to the manufacturer’s manual (Kit R, U-030, amaxa) with minor modifications. 1.0 × 10⁶ cells were nucleofected with 6 picomoles/sample control or Ucp2 siRNA (Dharmacon Ucp2 smart pool) and plated on 3 wells of the 6-well plate. 24 hours after nucleofection, the cells were trypsinized and 2.0 × 10⁴ cells were replated in the 12-well plate. 2 days after nucleofection, the cells were incubated with 1.0 × 10⁶ TAMRA-stained apoptotic SCI or Jurkat cells in DMEM supplemented with 10 % FBS and 1 % penicillin-streptomycin-glutamine at 37 °C for desired times.

Measurement of mitochondrial membrane potential

To detect mitochondrial membrane potential, cells treated with the indicated experimental conditions were stained with Mitotracker Deep Red FM (invitrogen), whose accumulation in mitochondria is dependent on mitochondrial membrane potential. Mitochondrial membrane potential was also measured using TMRE (invitrogen) or a dual emission potential-sensitive probe, JC-1 dye (SIGMA-Aldrich), according to the manufacturer’s protocol. For TMRE staining, cells were incubated with 50 nM of TMRE in culture medium in the 5 % CO₂ incubator at 37°C for 30 minutes and fluorescence coming from cells was measured by flow cytometry. For JC-1 staining, cells in a 24-well plate were incubated with 1 ml of mitochondrial staining solution containing the 50 % medium used for cell growth and 2.5 µg/ml of JC-1 dye in the 5 % CO₂ incubator at 37 °C for 20 minutes. The cells were then washed twice with warm PBS and trypsinized. Overflow of green fluorescent signal to red fluorescence was compensated and the intensity of red fluorescence coming from JC-1 aggregates was detected using the FACSCalibur instrument (Becton Dickinson).

Measurement of intracellular ATP

Intracellular ATP levels were measured using CellTiter-Glo Luminescent Cell Viability Assay kit (Promega) according to the manufacture’s manual. Cells were treated with experimental conditions and trypsinized. 2.5 × 10⁴ cells/50 µl were added into a well of the 96-well plate. After that 50 µl of CellTiter-Glo Reagent was added to the well containing the cells. The cells were incubated on orbital shaker for 2 minutes and followed by incubation at room temperature for 10 minutes. Luminescence was measured using MicroBeta TriLux (EG&G WALLAC).

Analysis of cell clearance in the thymus

Five to six week-old wild type and Ucp2 deficient mice were intraperitoneally injected with 300 µl of PBS containing 250 µg dexamethasone dissolved in ethanol (10 mg/ml). 6 hours after injection, the mice were sacrificed, and thymi were extracted and smashed. For quantification of thymic cellularity, thymocytes were resuspended with HBSS supplemented with 5 % FBS. The cells were diluted and mixed with 50µl of quantification beads (Spherotech Inc). The mixture was subjected to flow cytometry analysis to allow for the quantification thymocytes. To monitor the presence of TUNEL positive uncleared apoptotic cells, 6 hours after injection, the thymi from the mice were extracted and embedded in O.C.T. compound. The O.C.T compound embedded thymi then were frozen in liquid nitrogen. After that, thymic cryo-sections on glass slides were stained using the In Situ Cell Death Red kit according to manufacture’s instructions (Roche). Sections were mounted with Prolong Antifade + DAPI medium and analyzed by Axio imager 2 with Apotome (ZEISS). To monitor ex vivo apoptosis of thymocytes, 1.0 × 10⁷ thymocytes from 5~6-week-old Ucp2 wild type and deficient littermate mice were incubated with 50 µM dexamathasone at 37 °C for 4 or 6 hours. The cells were washed with PBS once and then stained with annexin V and propidium iodide (Annexin V: FITC apoptosis detection kit II, BD).

Cell migration assay

2 × 10⁶ cells/ml THP-1 cells were resuspended with RPMI containing 5 % FBS, 1 % penicillin, streptomycin and glutamine and 10 mM HEPES and were pre-incubated with the desired concentration of FCCP for 1 hour. 500 µl of apoptotic supernatant or MCP-1 (25 ng/ml) was added to lower chamber of a 24-well plate with 5 µm pore-size transwell (5 µm pore size, Corning) and 100 µl of cells (2 × 10⁶ cells/ml) was loaded on the upper chamber of the plate and then incubated at 37°C for 1 hour. The percentage of migrated cells was determined by FACS with 5.1-µm AccuCount beads and compared to input cells. For BMDMs migration, 4.0 × 10⁵ cells/ml were resuspended in RPMI containing 1 % BSA, 1 % penicillin, streptomycin and glutamine and 10 mM HEPES. 500 µl of apoptotic supernatant or Sdf-1 (50 ng/ml) was added to lower chamber of the 24-well transwell plate and 100 µl of BMDMs was added on upper chamber and incubated at 37°C for 3 hours. The number of BMDMs migrated on the underside of the membrane was determined by Diff-Quick staining and counted using microscopy.

Thymus staining

Tissue sections were blocked for 15 min with 2% normal goat serum and then were stained for 60 min at 25 °C with PE-conjugated F4/80 (eBioscience) and FITC-conjugated anti pancytokeratin (Sigma). After incubation, slides were washed three times in PBS and then mounted with GelMount (Molecular Probes, invitrogen).

Fatty acid oxidation (FAO) assay

Phagocytes were seeded in a 12-well plate at a density of 2 × 10⁵ cells per well on the day prior to the FAO assay. Next day, the phagocytes were incubated with 2 × 10⁷ apoptotic thymocytes or 2 µm carboxylate-modified beads for the desired times. After incubation, each well was washed five times with PBS. Rates of FAO were determined over the next two hours as follows: A 0.2 mL PCR tube with 50 µL 1M NaOH was leaned diagonally in each well prior to addition of 300 µL FAO buffer (Krebs Ringer phosphate buffer pH 7.4 containing 1% fatty acid free BSA (Sigma Aldrich), 5 mM glucose (Sigma Aldrich), 125 µM palmitate (Sigma Aldrich), 1mM carnitine (Sigma Aldrich), and 6 µCi 1-¹⁴C-palmitate (Perkin Elmer)). The wells were immediately sealed with masking tape and incubated at 37 °C for 2 hours before 100 µL of 2 M perchloric acid was injected through the tape to stop the reaction and release ¹⁴CO₂ trapped as bicarbonate. The plate was incubated for 2 hours at room temperature to trap the released ¹⁴CO₂ as bicarbonate in NaOH. The tape was removed and NaOH from each tube was diluted with 5 mL scintillant and counted with a Beckman LS6500 scintillation counter. Partially oxidized metabolites of ¹⁴C-palmitate were recovered from the FAO buffer precipitate by extraction in chloroform:methanol (2:1). Acid soluble ¹⁴C-metabolites were recovered from the aqueous phase, diluted in scintillant, and counted. Rates of total FAO were determined by summing ¹⁴CO₂ and ¹⁴C labeled acid soluble metabolite disintegration rates and normalizing to the specific activity of the FAO buffer, cell number, and the duration of incubation.

Measurement of superoxide

LR73s were plated at a density of 1.0 × 10⁵ cells/well in a 12-well plate in alpha-MEM supplemented with 10 % FBS and 1 % penicillin, streptomycin and glutamine. The cells were treated with indicated drugs for four hours and incubated with 2 µM MitoSox (invitrogen) for 30 minutes. The cells were trypsinized and analyzed by flow cytometry.

Testicular Torsion

This work was conducted in accordance with the Guiding Principals of the Care and Use of Research Animals promulgated by the Society for the Study of Reproduction. Adult male C57BL/6 mice were anesthetized with an intraperitoneal injection of 0.01 mg/g sodium pentobarbital and the testis was exteriorized through a low midline laparotomy, the gubernaculum was divided, and the testis was freed from the epididymo-testicular membrane. The testis was rotated 720° for 2 hr, during which time it remained in the abdomen with a closed incision. Following the 2 hr torsion, the incision was reopened, the testis was counter-rotated to the natural position, the gubernaculum was rejoined, and the testis was reinserted into the scrotum via the inguinal canal. Testes were examined at the time of repair for the degree of ischemia and reperfusion. Sham operated animals were treated identically except that upon completion of the torsion maneuver the testis was immediately counter-rotated.

Statistical analysis

Data are shown as the mean ± standard deviation (SD). For analysis of statistical difference of experiments involving two groups, Student’s two-tailed t test was applied. A one-way ANOVA was applied for statistical analysis of three or more groups. Significance was defined when p values were < 0.05.

Supplementary Material

Click here to view.^{(1.1M, pdf)}

Acknowledgments

We thank Dr. Martin Schwarts and the members of the Ravichandran lab for helpful discussions. This work was supported by grants from the NIGMS (to K.S.R.), ARRA funding from the Eunice Kennedy Shriver National Institutes of Child Health and Human Development (to J.J.L. and K.S.R.), and by funding for the Center for Cell Clearance. K.S.R. is a Bill Benter Senior Fellow of the American Asthma Foundation. J.M.K. is supported by an American Heart Association Award.

References

1. Elliott MR, Ravichandran KS. Clearance of apoptotic cells: implications in health and disease. J Cell Biol. 2010;189:1059–1070. [PMC free article] [PubMed]

2. Henson PM, Hume DA. Apoptotic cell removal in development and tissue homeostasis. Trends Immunol. 2006;27:244–250. [PubMed]

3. Gregory CD, Pound JD. Cell death in the neighbourhood: direct microenvironmental effects of apoptosis in normal and neoplastic tissues. J Pathol. 2011;223:178–195. [PubMed]

4. Munoz LE, Lauber K, Schiller M, Manfredi AA, Herrmann M. The role of defective clearance of apoptotic cells in systemic autoimmunity. Nat Rev Rheumatol. 2010;6:280–289. [PubMed]

5. Tabas I. Macrophage death and defective inflammation resolution in atherosclerosis. Nat Rev Immunol. 10:36–46. [PMC free article] [PubMed]

6. Affourtit C, Crichton PG, Parker N, Brand MD. Novel uncoupling proteins. Novartis Found Symp. 2007;287:70–80. discussion 80–91. [PubMed]

7. Esashi E, Sekiguchi T, Ito H, Koyasu S, Miyajima A. Cutting Edge: A possible role for CD4+ thymic macrophages as professional scavengers of apoptotic thymocytes. J Immunol. 2003;171:2773–2777. [PubMed]

8. Miyanishi M, et al. Identification of Tim4 as a phosphatidylserine receptor. Nature. 2007;450:435–439. [PubMed]

9. N AG, et al. Apoptotic cells promote their own clearance and immune tolerance through activation of the nuclear receptor LXR. Immunity. 2009;31:245–258. [PMC free article] [PubMed]

10. Krauss S, Zhang CY, Lowell BB. A significant portion of mitochondrial proton leak in intact thymocytes depends on expression of UCP2. Proc Natl Acad Sci U S A. 2002;99:118–122. [PubMed]

11. Krauss S, Zhang CY, Lowell BB. The mitochondrial uncoupling-protein homologues. Nat Rev Mol Cell Biol. 2005;6:248–261. [PubMed]

12. Fleury C, et al. Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat Genet. 1997;15:269–272. [PubMed]

13. Arsenijevic D, et al. Disruption of the uncoupling protein-2 gene in mice reveals a role in immunity and reactive oxygen species production. Nat Genet. 2000;26:435–439. [PubMed]

14. Blanc J, et al. Protective role of uncoupling protein 2 in atherosclerosis. Circulation. 2003;107:388–390. [PubMed]

15. Fisler JS, Warden CH. Uncoupling proteins, dietary fat and the metabolic syndrome. Nutr Metab (Lond) 2006;3:38. [PMC free article] [PubMed]

16. Moukdar F, et al. Reduced antioxidant capacity and diet-induced atherosclerosis in uncoupling protein-2-deficient mice. J Lipid Res. 2009;50:59–70. [PubMed]

17. Reers M, Smith TW, Chen LB. J-aggregate formation of a carbocyanine as a quantitative fluorescent indicator of membrane potential. Biochemistry. 1991;30:4480–4486. [PubMed]

18. Elliott MR, et al. Unexpected requirement for ELMO1 in clearance of apoptotic germ cells in vivo. Nature. 2010;467:333–337. [PubMed]

19. Park D, et al. BAI1 is an engulfment receptor for apoptotic cells upstream of the ELMO/Dock180/Rac module. Nature. 2007;450:430–434. [PubMed]

20. Lowell BB, Spiegelman BM. Towards a molecular understanding of adaptive thermogenesis. Nature. 2000;404:652–660. [PubMed]

21. Enerback S, et al. Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese. Nature. 1997;387:90–94. [PubMed]

22. Lugli E, et al. Characterization of cells with different mitochondrial membrane potential during apoptosis. Cytometry A. 2005;68:28–35. [PubMed]

23. Zhang CY, et al. Genipin inhibits UCP2-mediated proton leak and acutely reverses obesity- and high glucose-induced beta cell dysfunction in isolated pancreatic islets. Cell Metab. 2006;3:417–427. [PubMed]

24. Echtay KS, Winkler E, Bienengraeber M, Klingenberg M. Site-directed mutagenesis identifies residues in uncoupling protein (UCP1) involved in three different functions. Biochemistry. 2000;39:3311–3317. [PubMed]

25. Urbankova E, Hanak P, Skobisova E, Ruzicka M, Jezek P. Substitutional mutations in the uncoupling protein-specific sequences of mitochondrial uncoupling protein UCP1 lead to the reduction of fatty acid-induced H+ uniport. Int J Biochem Cell Biol. 2003;35:212–220. [PubMed]

26. Criscuolo F, Mozo J, Hurtaud C, Nubel T, Bouillaud F. UCP2, UCP3, avUCP, what do they do when proton transport is not stimulated? Possible relevance to pyruvate and glutamine metabolism. Biochim Biophys Acta. 2006;1757:1284–1291. [PubMed]

27. Buckler KJ, Vaughan-Jones RD. Effects of mitochondrial uncouplers on intracellular calcium, pH and membrane potential in rat carotid body type I cells. J Physiol. 1998;513(Pt 3):819–833. [PubMed]

28. Sokal A, Bartosz G. Uncouplers of mitochondrial oxidative phosphorylation are not substrates of the erythrocyte glutathione-S-conjugate pump. Arch Biochem Biophys. 1998;349:113–121. [PubMed]

29. Park D, Hochreiter-Hufford A, Ravichandran KS. The phosphatidylserine receptor TIM-4 does not mediate direct signaling. Curr Biol. 2009;19:346–351. [PubMed]

30. Kobayashi N, et al. TIM-1 and TIM-4 glycoproteins bind phosphatidylserine and mediate uptake of apoptotic cells. Immunity. 2007;27:927–940. [PMC free article] [PubMed]

31. Elliott MR, et al. Nucleotides released by apoptotic cells act as a find-me signal to promote phagocytic clearance. Nature. 2009;461:282–286. [PMC free article] [PubMed]

32. Gregory C. Cell biology: Sent by the scent of death. Nature. 2009;461:181–182. [PubMed]

33. Lauber K, Blumenthal SG, Waibel M, Wesselborg S. Clearance of apoptotic cells: getting rid of the corpses. Mol Cell. 2004;14:277–287. [PubMed]

34. Ravichandran KS, Lorenz U. Engulfment of apoptotic cells: signals for a good meal. Nat Rev Immunol. 2007;7:964–974. [PubMed]

35. Bereiter-Hahn J, Jendrach M. Mitochondrial dynamics. Int Rev Cell Mol Biol. 284:1–65. [PubMed]

36. Ferreira IL, Resende R, Ferreiro E, Rego AC, Pereira CF. Multiple Defects in Energy Metabolism in Alzheimer's Disease. Curr Drug Targets [PubMed]

37. Yoon Y, Galloway CA, Jhun BS, Yu T. Mitochondrial Dynamics in Diabetes. Antioxid Redox Signal

38. Azzu V, Brand MD. The on-off switches of the mitochondrial uncoupling proteins. Trends Biochem Sci. 35:298–307. [PubMed]

39. Emre Y, Nubel T. Uncoupling protein UCP2: when mitochondrial activity meets immunity. FEBS Lett. 584:1437–1442. [PubMed]

40. Tseng YH, Cypess AM, Kahn CR. Cellular bioenergetics as a target for obesity therapy. Nat Rev Drug Discov. 9:465–482. [PMC free article] [PubMed]

41. Andrews ZB, et al. Ghrelin promotes and protects nigrostriatal dopamine function via a UCP2-dependent mitochondrial mechanism. J Neurosci. 2009;29:14057–14065. [PMC free article] [PubMed]

42. Andrews ZB, et al. UCP2 mediates ghrelin's action on NPY/AgRP neurons by lowering free radicals. Nature. 2008;454:846–851. [PubMed]

43. Pi J, Collins S. Reactive oxygen species and uncoupling protein 2 in pancreatic beta-cell function. Diabetes Obes Metab. 12(Suppl 2):141–148. [PubMed]

44. Das S, et al. Brain angiogenesis inhibitor 1 (BAI1) is a pattern recognition receptor that mediates macrophage binding and engulfment of Gram-negative bacteria. Proc Natl Acad Sci U S A. 2011;108:2136–2141. [PubMed]