Zn2+ is an important structural and functional component in many cellular proteins and enzymes. As such, Zn2+ levels are normally tightly regulated, limiting the extent of cytosolic labile (or free) Zn2+ concentrations [1,2]. For example, levels of free Zn2+ are several orders of magnitude less than that of Ca2+ . Zn2+ may act as a cellular messenger in physiological and cytotoxic signaling, and the changes in Zn2+ homeostasis have a fundamental effect in cell function [4,5]. Many studies have shown the accumulation of excessive Zn2+ to precede cell death or neurodegeneration in response to cytotoxic stress [6,7]. To characterize Zn2+-mediated signaling pathways or Zn2+-induced cytotoxicity, it is important to determine the source(s) of intracellular free Zn2+ in response to specific stimuli or injury.
The endoplasmic reticulum (ER) is an intracellular organelle that has been shown to sequester Ca2+ from the cytosol by means of sarcoplasmic/endoplasmic reticulum Ca2+-ATPase (SERCA) or so-called endoplasmic Ca2+ pump . This sequestered Ca2+ can be released into the cytosol upon a variety of stimuli including inositol 1,4,5-trisphosphate (IP3). It is IP3 that mobilized Ca2+ from the ER Ca2+ store following interaction with specific IP3 receptors (IP3R). A commonly used tool in studying Ca2+ homeostasis is thapsigargin, a plant derived compound that specifically inhibits SERCA activity . By blocking the ability of the cell to pump Ca2+ into the ER, thapsigargin causes these stores to become depleted and thereby raise the cytosolic Ca2+ concentration.
While the mechanisms responsible for regulating Zn2+ homeostasis are not well established, available data support that, like Ca2+, intracellular Zn2+ levels are determined by the interaction of membrane Zn2+ transporters and cytoplasmic Zn2+ buffers [4,10]. The present study investigates the intracellular source of free Zn2+, particularly, if thapsigargin can trigger the release of Zn2+. This possibility is supported by recent evidence that Zn2+ can be released from intracellular sources upon stimulation [11 - 13]. Our results show that Zn2+ is released from thapsigargin-sensitive and IP3R-mediated stores.
Primary Cell Culture
Pregnant Sprague-Dawley rats (E17-E18) were anaesthetized with CO2 and the fetuses were removed and placed in ice-cold Hank's Balanced Salt Solution without Ca2+ or Mg2+ (HBSS). The brains of fetuses were removed and placed into cold HBSS for further dissection. Using a dissecting microscope and blunt dissection, the meninges were gently separated away. The cerebral cortex was then removed and each cortical hemisphere was cut into four pieces and trypsinized in HBSS at 37°C. Following trypsinization, cells were separated by trituration through the opening of a fire polished Pasteur pipette. The suspensions were then passed through a 70 μm cell strainer. The dissociated cells were added to the bottom of 35mm glass-bottomed petri dishes previously coated with polyethyleneimine (50% solution, Sigma, St. Louis) diluted 1:1000 in borate buffer. The cortical neurons were then allowed to attach to the surface at 37°C, 5% CO2 in 2 ml of MEM solution (Gibco, BRL) supplemented with 10% (v/v) heat-inactivated fetal bovine serum. After 3-6 hrs, solutions were replaced with fresh supplemented MEM which was later replaced (24 hrs) with Neurobasal medium (Gibco, BRL) supplemented with 2% B-27.
All imaging experiments were performed in HEPES medium containing the following (in mM): 130 NaCl, 5 KCl, 8 MgSO4, 1 Na2HPO4, 25 Glucose, 20 HEPES, 1 Na-Pyruvate; pH adjusted to 7.4. Cortical cells grown in glass-bottomed petri dishes were washed with fresh HEPES medium. The cells were then incubated at 37°C for 30 min with ER-Tracker Red (Molecular Probes, Carlsbad, CA) and Newport Green (Molecular Probes, Carlsbad, CA) or ZinPyr-1 (Neurobiotex, Galveston, TX). Cells were incubated with 10 μM of the specified fluorescent Zn2+ indicator either alone or in conjunction with ER-Tracker red for 30 min then were washed 3x with fresh HEPES medium and placed into a custom 35mm stage adapter and continuously perfused with fresh HEPES medium on the stage of a Zeiss LSM 510 (confocal) microscope. Cultures were examined using a Plan-Neofluar 100x/1.3 NA oil immersion objective. For ER-Tracker Red excitation was done with a HeNe Laser line of 543nm and an LP560nm emission filter. Newport Green and ZinPyr-1 were imaged using an Argon Laser line of 488nm for excitation and a BP505-550nm emission filter. Separate fluorescent channels were employed for each indicator, and channels were scanned sequentially to minimize crosstalk. Cells were imaged by serial z-scans progressively from bottom to top, in increments of 500 nm. Colocalization of ER-Tracker Red and Newport Green or ZinPyr-1 was measured using Zeiss software . Colocalization was determined by Pearson's correlation coefficient and considered significant when (p ≤ 0.05). Time series measurements of fluorescence intensity were done with image capture at 10 sec intervals, and changes in intensity were measured using Zeiss LSM 510 image analysis software. Fluorescence measurements were background subtracted, normalized to starting values, and expressed as F/Fo.
Caged IP3 experiment
To directly activate IP3Rs we used the membrane-permeable UV light-sensitive caged IP3 analogue, ci-IP3/PM (D-2,3-O-isopropylidene-6-O-(2-nitro-4,5-dimethoxy) benzyl-myo-inositol 1,4,5-trisphosphate-hexakis(propionoxymethyl)ester (SiChem. Bremen, Germany) [15,16]. Cells in brain hippocampal slices were simultaneously loaded by incubation with caged IP3 and Newport Green for 30 min at 37°C, then washed and incubated for an additional 30 min to allow for complete de-esterification.
IP3 was photoreleased by flashes of 364 nm light focused uniformly throughout the field of view.
Thapsigargin (Molecular Probes, Carlsbad, CA), the IP3K inhibitor N2-(m-trifluorobenzyl),N6-(p-nitrobenzyl)purine (Calbiochem, Cat. No. 406170), N,N,N',N'-tetrakis (2 pyridylmethyl) ethylenediamine (TPEN) (Molecular Probes, Carlsbad, CA), the Ins(1,4,5)P3 receptor blocker 2-aminoethoxydiphenyl-borate (2-APB), were applied by bath application in HEPES medium.
Intracellular Regions of Elevated Zn2+ Co-localize with the Endoplasmic Reticulum
Cells labeled with intracellular fluorescent Zn2+ indicators and examined under basal conditions showed consistent regions of elevated fluorescent intensity in the soma and processes, and particularly in a region that was identified as the endoplasmic reticulum by a fluorescent marker for the organelle. The elevated levels of Zn2+ were seen to represent labile or free Zn2+ because they were sensitive to both low and high affinity fluorescent Zn2+ indicators Newport Green (KDZn2+ ≈ 10-6M) (Figure 1A) and ZinPyr-1(KDZn2+ ~ 10-9M) (Figure 1B). Before imaging, cells were washed three times with fresh medium to remove excessive fluorescent residues. Another reason for using ZinPyr-1 is that, unlike AM forms of fluorescence indicators, it is highly lipophilic and remains in the organelle, being sequestrated with zinc. These indicators are essentially insensitive to Ca2+ and their fluorescence to Zn2+ are not altered in the presence of Ca2+ [17,18]. Fluorescence was sensitive to quenching by the membrane permeable Zn2+-chelator TPEN (KDZn2+ ~ 10-17 M) (Figure 2). The same cells were also loaded with ER-tracker Red, a marker for the ER, to determine the localization of the intracellular fluorescent Zn2+ indicators. We performed serial z-scans using confocal microscopy of cortical neurons loaded with ER-tracker Red and either Newport Green or ZinPyr-1 (Figure 1A, B). Both fluorescent Zn2+ indicators showed significant colocalization with the ER Tracker Red. Colocalization was abolished upon exposure to 10 μM TPEN (Figure 2), indicating that the local Zn2+ fluorescence represented free Zn2+ in the basal condition and were located within the lumen of the ER.
Thapsigargin-Induced Zn2+ Release
To assess the overall dynamics of Zn2+ release from thapsigargin-sensitive stores, cells incubated with Newport Green were exposed to thapsigargin, which inhibits SERCA pump. The application of thapsigargin depletes the ER Ca2+ stores in cells and raises cytosolic Ca2+ concentration. The experiments above suggested that Zn2+ may be also transported into and sequestered in ER. If this was true, the accumulation of Zn2+ by blocking SERCA with thapsigargin should also produce Zn2+ signals. Indeed, exposure to thapsigargin resulted in a gradual increase of cytosolic or intracellular Zn2+ (Figure 3), which was sensitive to the Zn2+ chelator TPEN (Figure 4).
In the next two tests we determined that the source of this thapsigargin-induced Zn2+ increases. One possible source is the influx of extracellular Zn2+. To remove extracellular Zn2+, we applied a membrane impermeable Zn2+ chelator CaEDTA. As shown in figure 4, thapsigargin-induced Zn2+ rises were unchanged in the presence of CaEDTA (1 mM). Next, we examined the contribution of extracellular Ca2+ on thapsigargin-induced elevation of intracellular Zn2+. We found that the removal of extracellular Ca2+ had no effect on the thapsigargin-induced Zn2+ rises (Figure 4). Taken together, these results indicated that thapsigargin-induced increases in intracellular Zn2+ were not dependent on either extracellular Zn2+or extracellular Ca2+, and were entirely of intracellular origin.
IP3-Induced Intracellular Zn2+ Release
The endoplasmic reticulum is a well established site of intracellular Ca2+ storage and release. IP3 can trigger the release of Ca2+ from intracellular stores by binding to and activating its receptor (IP3R) located on regions of the ER. When IP3 binds to and activates IP3Rs, the channel portion of the receptor opens and Ca2+ is released from the ER to the cytosol. The experiment was therefore performed utilizing ci-IP3/PM, a cell-permeable form of caged IP3 to directly induce Zn2+ release [15,16,19 - 21]. In neurons loaded with the Zn2+ fluorophore Newport Green and caged IP3, IP3 uncaging resulted in a rapid increase in intracellular Zn2+, which persisted for 30 s and followed by a roughly exponential decay (Figure 5A &5B). This experiment demonstrated that the Zn2+ response was due to the activation of IP3 receptors.
A significant route of IP3 metabolism is the conversion of IP3 into inositol (1,3,4,5)-tetrakisphosphate (Ins(1,3,4,5)P4) by the enzyme Ins(1,4,5)P3 3-kinase (IP3-3Kinase or IP3-3K) . Inhibition of the IP3-3K has been shown to elevate intracellular levels of IP3 by halting its conversion into Ins(1,3,4,5)P4 . Here, to examine the effects of IP3 signaling on intracellular Zn2+ dynamics, N2-(m-trifluorobenzyl),N6-(p-nitrobenzyl)purine, a membrane-permeable inhibitor of IP3-3K was employed to confirm the results observed using the caged IP3. Upon bath application of the inhibitor, Newport Green fluorescence showed a gradual, significant increase (Figure 5C &5D), supporting an IP3 mediated process involved in the release of Zn2+.
The major findings of the present study are the following: Neuronal cells maintain a substantial concentration of Zn2+ in ER-like storage, and Zn2+ is released into the cytosol in a thapsigargin- and IP3-sensitive manner. These findings suggest a new model of intracellular Zn2+ homeostasis where cellular organelles like the ER act as sites of intracellular Zn2+ storage.
Available data support that intracellular Zn2+ levels can be determined by the interaction of membrane Zn2+ transporters and cytoplasmic Zn2+ buffers [4,10]. In eukaryotic cells the concentration of intracellular Zn2+ has been found to be in the range of a few hundred micromolars . The vast majority of this cellular Zn2+ is, however, protein bound or sequestered into organelles, which results in free cytosolic Zn2+ concentrations being in the picomolar to nanomolar range [10,24,25]. This is consistent with our observations that there is a sharp contrast in Zn2+ fluorescence between ER-like lumen and cytosolic space (Figure 1). The significant concentration of free (labile) Zn2+ present in ER-like lumen suggests the existence of a high intraluminal Zn2+ sequestering activity.
The present study shows that Zn2+ is released from thapsigargin-sensitive and IP3R-mediated stores (Figure 3 &5). Thapsigargin, as expected, also induced a Ca2+ transient measured with a fluorescent Ca2+ indicator (data not shown but see [26 - 28]). Collectively, we suggest that Ca2+ is not the only metal ion that is sequestered in the ER. Zn2+ is released alongside of Ca2+ upon thapsigargin stimulation. We show further that the source of thapsigargin-induced elevation in intracellular Zn2+ is of intracellular origin. The elevation of Zn2+ is independent from either extracellular Ca2+ or extracellular Zn2+. These results indicate that Zn2+ homeostasis, like Ca2+ homeostasis, is controlled by IP3Rs (Figure 5) that may gate Zn2+ into the cytoplasm, and by thapsigargin sensitive ATPase activity that pumps Zn2+ from the cytoplasm into the ER.
Within the ER, it is known that Ca2+ is buffered by the abundant luminal resident chaperone protein calreticulin which binds Ca2+. Although calreticulin was first identified as a Ca2+ binding protein , this protein is multifunctional  and binds other ions including Zn2+ with multiple binding sites [31 - 34]. Zn2+ also binds with several other luminal proteins . Just like Ca2+, recent work indicates that mitochondria take up cytosolic Zn2+ and that Zn2+ accumulation leads to a loss of mitochondrial membrane potential (see review ). There are reports suggesting that thapsigargin/IP3 regulate mitochondrial Ca2+ signaling and function . It remains to be studied how thapsigargin and IP3 induced Zn2+ release affect mitochondrial function.
While the mechanism(s) that govern Zn2+ trafficking remain elusive, there is little doubt that the intracellular free Zn2+ level must be maintained within a physiological limit. Zn2+ has been shown to activate a number of protein kinases such as protein kinase C, CaMKII, TrkB, Ras and MAP kinase [38 - 43]. On the other hand, abnormal levels of Zn2+ may lead to either Zn2+-induced toxicity or Zn2+ deficiency-induced apoptosis . Therefore, thapsigargin sensitive storage or the ER may function as a source of intracellular free Zn2+ in response to stimuli, and is likely to play an important role in the regulation of intracellular levels of Zn2+.
The authors declare that they have no competing interests.
CS carried out the fluorescence imaging experiments, analysed data, and participated in the experimental design and the preparation of manuscript. YL conceived of the study, and participated in its design, and drafted and prepared manuscript, and coordination. All authors read and approved the final manuscript.