Redistribution of Labile Plasma Zinc During Mild Surgical Stress in the Rat

Abstract

Zinc is an essential trace element and co-factor for many cellular processes. Uptake of Zn²⁺ in peripheral tissues depends on its total content in the circulation, and on mechanisms facilitating delivery to tissues in its labile form. Understanding mechanisms of Zn²⁺ delivery has been hindered by the absence of techniques to detect labile Zn²⁺ in the circulation. In this study, we report the use of the fluorescent zinc-binding dye, ZnAF-2, to detect changes in labile Zn²⁺ in the circulating plasma of the rat under standardized conditions, including exogenous infusions to raise plasma Zn²⁺, and infusion of the chelator, citrate, to lower labile Zn²⁺ in the plasma without altering total Zn2+ content. In a model of mild surgical stress (unilateral femoral arterial ligation), plasma levels of total and labile Zn²⁺ decreased significantly 24 hours following operation. Ultrafiltration of plasma into high and low molecular weight macromolecule fractionations indicated that binding capacity of zinc in the high molecular weight fraction is impaired for the entire 24 hour interval following induction of mild surgical stress. Affinity of the filtrate fraction was rapidly and reversibly responsive to anesthesia alone, decreasing significantly at 4 hours and recovering at 24 hours; in animals subjected to moderate surgical stress this responsiveness was lost. These are the first reported measurements of labile Zn²⁺ in the circulation in any form of mild systemic stress. Zinc undergoes substantial redistribution in the plasma, response to surgical stress, leading to increased availability in lower molecular weight fractions and in its labile form.

Introduction

Zinc is an essential trace element for all living cells and is a co-factor for many cellular processes including protein synthesis, energy metabolism, nucleic acid synthesis, gene transcription, and programmed cell death (1,2). A wealth of experimental (3,4,5,6) and clinical (7,8,9) data indicate that total levels of Zn²⁺ in the circulation are decreased in a variety of chronic and acute conditions associated with impaired immune response. The groundbreaking work of Prasad, beginning in the 1960s demonstrated the biological syndrome of chronic zinc deficiency in humans (10,11,12), and clinical response to zinc supplementation (13,14). Subsequent research has also demonstrated the value of zinc supplementation in other zinc-avid conditions such as bone marrow transplantation (15), uremia (16), and neurodegenerative disease (17).

To date most zinc assays used in biological research have reported total zinc content of tissue or fluids. The total content of Zn²⁺ in the plasma reflects components that are bound to plasma proteins such as albumin and alpha-2 macroglobulin (18), constituting more than 99% of the total content in the plasma. The component that is loosely bound or free is designated as “labile” Zn²⁺. It is this component that is accessible to meet the requirements for key signaling and phagocytic functions (19, 20, 21) of cells in the circulation (22, 23), including leukocytes (24) and endothelium (2). These considerations indicate that ongoing availability of labile Zn²⁺ in the circulation and its delivery to peripheral tissues is a vital factor in the cellular response to shock, injury and infection. As with calcium, a divalent cation with parallel activities in biology, the availability of free zinc may not correlate directly to the total circulating concentration.

Investigation of the mechanisms of delivery has been impeded by the absence of a method to detect free Zn²⁺ in the circulation. This report summarizes our experiments using ZnAF-2, a fluorescent dye that binds zinc selectively, as a reporter for nanomolar concentrations of free zinc concentration in the plasma. The present study was performed to investigate the concentration of circulating total and labile Zn²⁺, in a rat model without prior zinc deprivation under conditions of moderate surgical stress. Elimination of malnutrition and zinc deficiency from the model enables investigation of labile zinc delivery to tissues in response to stress alone, in order to study the alteration of zinc availability that may be physiologically adaptive, as opposed to the effects of zinc depletion. To our knowledge these are the first reported efforts to obtain measurements of changes in plasma labile Zn²⁺ in such a clinical model.

MATERIALS AND METHODS

Reagents and solutions

Unless otherwise noted, all reagents were from Sigma-Aldrich (St. Louis, MO). ZnAF-2 was purchased from Axxora (San Diego, CA). Fluozin-3 was purchased from Invitrogen/Molecular Probes (Carlsbad, CA). Ringer’s solutions used for preliminary screening studies contained NaCl 145mM, KH₂PO₄ 2.5mM, MgSO4 1mM, HEPES 10mM, EGTA 0.3mM, pH 7.4, with Ca²⁺ and Zn²⁺ added to maintain Ca²⁺ at 1mM and Zn²⁺ at desired concentrations. Free concentrations for Ca²⁺ and Zn²⁺ in Ringer’s solutions containing chelators such as EGTA or citrate were calculated based on the internet-accessible WEBMAXCHELATOR (http://www.stanford.edu/~cpatton/webmaxc/webmaxcS.htm).

Experimental Animals and Surgical Procedures

Male Sprague Dawley rats weighing 300–350g were used for all experiments (Charles River Labs, Waltham, MA). For some studies, animals were purchased with cannulae (microrenethane MRE-040, external diameter 0.04 inches, internal diameter 0.025 inches) pre-placed in the femoral artery or vein several days prior to blood draw, in order to minimize acute stress. Rats were housed using standard animal care procedures (12:12 hour light-dark cycle, food and water ad libitum). All animal care and experimental procedures used were consistent with National Institutes of Health Animal Care and Use Guidelines and were approved by the Institutional Animal Care and Use Committee at Harvard Medical School. Rats were maintained under general anesthesia using pentobarbital intraperitoneal injection. Warming pads were used to maintain temperature.

For vascular access, the lower abdomen and groins were shaved and the femoral vessels were surgically exposed. Cannulation of artery or vein was performed with 24-gauge IV catheters (Angiocath, BD Medical, Sandy, Utah), accompanied by distal ligation using fine silk ligatures. The arterial catheter was used for invasive blood pressure measurement using a Spacelabs multichannel monitor (Spacelabs Healthcare, Issaquah, Washington). The catheter was flushed with normal saline to maintain patency and provide fluid boluses, and was also used for withdrawal of blood samples (1.5 ml at each time point) from the most proximal port (~10 cm from the catheter) to minimize dead space.

Screening of conditions and candidate reporters for measuring free Zn²⁺ in plasma

Preliminary studies were performed to identify optimal conditions for assay, and included exploration of the influence of simple salt solutions (100uM NaCl) containing different pH-buffers (10mM HEPES, Glycine, N-methyl-D-glucamine, or mono-/bi-sodium phosphate) on responsiveness of those reporters to Zn²⁺. The resulting solutions were plated in 96-well plates and fluorescence was measured in a Synergy 2 microplate reader (Biotek Inc, Winooski, VT). Since content of labile Zn²⁺ in the plasma has been predicted to reside in the nanomolar range or lower (25, 26, 27), Fluozin-3 and ZnAF-2 (50nM each) were selected for investigation based on reports of their high selectivity for zinc and dissociation constants in the low nanomolar range in aqueous solution (28, 29, 30). Among the buffers, HEPES was the only one that did not interfere with measurements of reporter fluorescence in simple aqueous solution. In contrast to anticoagulants that are also chelators (EDTA, citrate) of divalent cations, we observed that heparin does not interfere with measurements in HEPES-buffered Ringer’s if placed in dilutions of 1:50 or less (data not shown).

Standardization of conditions for plasma assay of labile Zn²⁺ and fractionation of samples

Plasma samples were collected and measured immediately, as preliminary experiments demonstrated increased variability of measurements in samples stored on ice. Blood specimens were collected from the femoral artery catheter in 1.5 mL aliquots and centrifuged at 10,000 ×g for 10 minutes. The plasma fraction was immediately assayed for labile zinc. Total zinc, total protein and albumin were assayed in batches at the end of the experiment.

In a subset of experiments, a 500 µl aliquot of plasma from each sample was filtered through a 10 kDa plasma filter (Amicon Ultra, Millipore, Billerica, MA) at 14,000 ×g for 15 minutes at 4°C. The filtrates were retained and concentrates were resuspended in zinc-free 18 MΩ Milli-Q water. For measurements of labile Zn²⁺, whole or fractionated samples of plasma (50µl/sample) were plated in duplicate or triplicate on black 96-well plates (NUNC, Thermo Scientific, Roskilde, Denmark). After an initial baseline fluorescence reading (ex. 485, em. 528), ZnAF-2 (1.5µl of a 1 µM solution in DMSO) was added to each well, for a final concentration of 30 nM. Samples were mixed by gentle shaking for 10 seconds, and fluorescence was re-read.

Total Plasma Zinc assay

Plasma total zinc was determined using the Quantichrom Zinc Assay kit (BioAssay Systems, Hayward CA), according to the manufacturer’s instructions. This zinc assay is a colorimetric assay based on zinc binding to a chromogen that reports at 425 nm. Results were read on the microplate reader.

Plasma Albumin assay

Plasma albumin was assayed using the BCP Albumin Assay Kit (BioAssay Systems, Hayward, CA) according to the manufacturer’s instructions, and results were read at 610 nm on the plate reader.

Statistics

Data were analyzed using standard statistical software (SigmaStat v3.5, Systat Inc, Chicago, IL). Continuous variables are expressed as mean plus or minus standard error. One-way ANOVA was used for statistical comparison of multiple groups, and Student’s t-test was used for before and after comparison within groups where appropriate.

RESULTS

Evaluation of fluorescent Zn²⁺ reporters for plasma assay

In order to assess the validity of using vital dyes for determination of zinc concentration in solution, we evaluated two candidate reporters, based on their known binding affinities for zinc in the nanomolar range: Fluozin-3 and ZnAF-2 (both in their free acid forms). Both reporters excite in the green component region of the spectrum (485nm), with emissions measured at 528 nm. When studied in simple HEPES-buffered Ringer’s solutions, good titration curves were obtained for Fluozin-3, confirming published reports that the K_d for Zn²⁺ (31, 32) lies in the low nanomolar range. When evaluated in Ringer’s solutions (Figure 1A), ZnAF-2 provided a reliable range of fluorescence responses (from F_min to F_max) in response to increases in [Zn²⁺] from ~0nM (Ringer’s + 500µM EGTA) to 32nM. When evaluated in plasma, the response curve maintained its dynamic range, that is, the ratio of the fluorescence minimum when all labile Zn²⁺ is chelated with excess EGTA, compared to the fluorescence maximum in the presence of excess Zn²⁺ (33) In HEPES-Ringer’s and plasma, this ratio is about 1:8. In addition, the curve shifted markedly to the right (Figure 1B), indicating that the dye is capable of monitoring changes in labile Zn²⁺ despite the presence of high affinity binding within the plasma. These observations accord with published estimates of labile zinc concentration in the plasma in the range of 2–10 nM.

Figure 1.

Titration curves for ZnAF-2 fluorescence measured with excitation 485nm and emission at 528nm. In Ringer’s solutions (A), the dye is responsive to changes in free concentration of Zn²⁺, over the range 0 nM to 32 nM, confirming the profound buffering capacity of the plasma for Zn²⁺. In rat plasma (B), the same range responsiveness is observed in rat plasma with all Zn²⁺ chelated [POINT X] or in samples supplemented with high concentrations of exogenous ZnCl₂, at levels up to 16 mM, indicating that responsiveness of the dye is preserved in plasma. In panel B the vertical dashed line indicates the fluorescence of untreated plasma minus background. This value, about 1400 RFU, corresponds with a concentration of about 1 nM of labile zinc as predicted by the Grynkiewicz fluorescence equation (33). The arrow in panel A marks the corresponding concentration, 1 nM protein-free Ringers solution. All data are summarized as mean relative fluorescence units (RFU) ± SD, N = 3 for all data points.

Of note, excitation of plasma at 485nm does yield background fluorescence. We further found that background fluorescence increases with Zn²⁺ content, increasing from about 10% of signal under baseline conditions to about 20% of signal when 100µM of Zn²⁺ is added to a plasma sample. As a result, all measurements require correction for background for each experimental condition, and this was done in the studies that are reported below. Of note, Figure 1A shows that the background fluorescence of dye in Ringers solution is about 4000 units, which was constant throughout our experiments. On the other hand, in plasma (Figure 1B) the difference between background fluorescence of dye in plasma and fluorescence in response to added zinc is much less. Because of the difficulty of independently calibrating the dye in a complex fluid such as plasma under different circumstances, it seemed most appropriate to summarize results by comparison with baseline measurements (normalization to a baseline level of 1.0). Extrapolations to quantitative measurements are provided in the discussion, below.

Sequential measurements in plasma during exogenous infusions of Zn²⁺ and citrate

We next determined whether the ZnAF-2 assay is capable of detecting controlled changes in the labile Zn²⁺ content of plasma. Rats underwent cannulation of both femoral veins and one femoral artery and were divided into three groups (n=4 each). In the first (control) group, only normal saline was infused in each of the venous catheters (1cc over 10 min). In the second group (Zn²⁺-infusion), a brief bolus infusion of ZnCl₂ was performed through one of the venous cannulae (0.4 mg/kg, 1cc over 10 min) while saline was administered through the other (1cc over 10 min). In order to demonstrate detection of transient changes in labile Zn²⁺ in plasma, a third group of animals received infusion of a Zn²⁺ bolus infusion through one cannula while receiving infusion of sodium citrate (30mg/kg in 1cc, over 10 min) through the other. Citrate is a moderate affinity chelator for divalent cations including Zn²⁺ (K_d for Zn²⁺ 18µM). Arterial blood samples were drawn before (time 0) and 5, 15, 30, and 60 minutes after infusion) for measurements of total and labile Zn²⁺ in plasma.

Measurements of total Zn²⁺ content in plasma, using a commercial colorimetric assay, are shown in Figure 2 and those of labile Zn²⁺ are shown in Figure 3. Among control animals, neither total nor labile Zn²⁺ levels were changed significantly over baseline. Among the animals who received infusion of Zn²⁺ alone, levels of total Zn²⁺ were elevated immediately following the infusion and remained elevated for the duration of study period, slowly returning toward baseline from a peak at 5 minutes. In these animals, increases in levels of labile Zn²⁺ were also detected, peaking at 40% to 50% above baseline at 10 to 20 minutes and then slowly declining thereafter. Among the animals receiving both Zn²⁺ and the divalent cation chelator sodium citrate, each infused through separate catheters, there was a similar rise and time course of elevation in levels of total Zn²⁺. In contrast to animals undergoing infusion of Zn²⁺ alone, initial increases in labile levels were observed before a sharp, transient downturn was detected when the metal and the chelator had time to intermix in the circulation.

Figure 2.

Total zinc measurements in whole plasma samples from rats undergoing infusions of saline (5ml over 20 min, symbol ●), exogenous zinc sulfate (0.4 mg/kg in 5 ml over 20 min, symbol ○), or exogenous zinc sulfate (0.4 mg/kg) and sodium citrate (30 mg/kg) through separate infusion catheters (symbol ▼). Data points are shown as mean ± SEM. * = p < 0.05 vs. control, N =4.

Figure 3.

Labile Zn²⁺ measurements in whole plasma samples, using ZnAF2 fluorescence, during infusions of saline, zinc sulfate or zinc sulfate and citrate (same animal groups as in Figure 2). Rats underwent infusions of saline (5ml over 20 min, symbol ●), exogenous zinc sulfate (0.4 mg/kg in 5ml over 20 min, symbol ○), or exogenous zinc sulfate (0.4 mg/kg) and sodium citrate (30 mg/kg) through separate infusion catheters (symbol ▼). Labile zinc rises sharply in the earliest time points after infusion, buffered transiently with the concurrent infusion of the chelator, citrate. Data points are shown as mean ± SEM. * = p < 0.05 vs. control, N =4.

Alterations in plasma Zn²⁺ levels during moderate surgical stress

To evaluate disturbances in circulating levels of labile Zn²⁺, we performed studies in a rat model of moderate surgical stress. Rats underwent pentobarbital anesthesia, skin incision, tissue dissection, unilateral femoral artery ligation and cannulation, and blood sampling at time points A (time 0), B (one hour following cannulation), and C (five hours following cannulation). The femoral artery was then decannulated and ligated, and rats were allowed to awaken. Twenty-four hours following the start of the initial surgical procedure, rats were re-anesthetized and the femoral artery was cannulated on the contralateral side for collection of a final blood sample (time point D). Rats were then euthanized by pentobarbital overdose and exsanguination. Rats that had been pre-cannulated in the femoral artery served as controls and were placed under anesthesia only using an identical protocol. Measurements were obtained in both groups (n=6 for each group).

Summarized in Figure 4 are changes in levels of total Zn²⁺ in plasma, in pre-cannulated control rats undergoing anesthesia alone (Figure 4A) and in rats acutely, but mildly stressed by anesthesia and femoral artery cannulation (Figure 4B). While some variations were observed over the 24 hours of study in the control group, there was a significant decline in total plasma content of Zn²⁺, in the stressed group at the 24 hour time point. Summarized in Figure 5 are changes in labile Zn²⁺. In control animals these were remarkably constant (Figure 5A) over the period of study, while in the stress group (Figure 5B) they decreased significantly in period D. In this 24 hr interval following initiation of study, significant alterations in plasma albumin -- a key serum binding protein for Zn²⁺-- were not observed in control or stressed groups (Figure 6). Thus, in period D, the ratios of total (Figure 7) Zn²⁺ to serum albumin were markedly decreased (p<0.05). These findings suggest that the decrease in total Zn²⁺ was not due to alterations in plasma concentration of binding proteins; rather, the decline in total Zn²⁺ appears to reflect unloading of Zn²⁺ due to changes in plasma protein binding capacity.

Figure 4.

Total zinc measurements in whole plasma samples from animals pre-cannulated to avoid acute surgical stress, that were subjected to anesthesia alone or animals subjected to anesthesia and a mild surgical stress. In control animals (A), there were not significant alterations throughout the 24 hour experimental period. In stressed animals (B), levels decreased progressively, becoming statistically different from baseline at 24 hours (Time Point D). Data points are shown as mean ± SEM. * = p < 0.05 vs. Time Point A, N = 6.

Figure 5.

Labile Zn²⁺ in whole plasma samples from animal groups undergoing anesthesia alone or mild surgical stress (same animal groups as in Figure 4). In control animals (A), no significant changes were observed during the 24 hr period of observation. Stressed animals (B) exhibited significant decrease; mean 40% below baseline at 24 hrs (Time Point D). Data points are shown as mean ± SEM. * = p < 0.05 vs. Time Point A, N = 6.

Figure 6.

Plasma albumin levels measured in plasma samples of animals undergoing anesthesia alone (A) or mild surgical stress (B); same animal groups as in Figure 4. No significant differences were observed over time in either group, confirming that stress in both groups was mild. The stable level of albumin concentration is consistent with mild stress, without significant acute phase response. Data points are shown as mean ± SEM. N = 6.

Figure 7.

Whole plasma total zinc:albumin ratio calculated from measurements reported in Figures 4 and 5. In the group subjected to mild surgical stress, there is a more than 50% decrease in the total zinc to albumin ratio in the Stress group at 24 hours (Time Point D). Data points are shown as mean ± SEM. * = p < 0.05 vs. Time Point A, N = 6.

We then explored potential alterations in distribution of Zn²⁺ among serum protein fractions. Plasma proteins were separated by filtration of whole plasma samples (0.5ml), into high (>10kDa) and low (<10kDa) fractions. Volume of each fraction was adjusted to 0.5ml, using 18 MΩ water, then assayed for total Zn²⁺ content (200µl aliquots). Shown in Figure 8A are measurements of total Zn²⁺ in plasma fractions filtrate (F, <10kDa) and concentrate (C, >10kDa) obtained from a group of animals (n=4) subjected to moderate surgical stress. Summarized in Figure 8B are calculations of the ratio of total Zn²⁺ in the F vs. C fractions, used as an index of redistribution between the two. The results indicate that changes in total Zn²⁺ content of plasma (Figure 4) largely reflect those in the large molecular weight fraction. In period D (24 hours), it appears that there is a transfer of Zn²⁺ content from the high molecular weight fraction to the low molecular weight fraction.

Figure 8.

A. Total zinc content in high and low molecular weight fractions obtained from plasma of animals undergoing mildly surgical stress. Plasma was separated by centrifugal filtration across a 10 kDa filter into concentrate (> 10 kDa, grey bars) and filtrate (< 10 kDa, black bars) fractions at each time point. B: Calculations for the ratio of Zn²⁺ content in the concentrate (>10kD fraction) to that in the low molecular weight fraction (<10kD fraction). The data indicate that, over the 24 hour time course of the experiment, even mild surgical stress induces a shift of Zn²⁺ content from the high- to the low-molecular weight fraction. Data points are shown as mean ± SEM. p = NS versus Time Point A. N = 4.

We then tested the hypothesis that surgical stress had altered the affinity of the larger and smaller protein fractions for Zn²⁺. Aliquots of reconstituted plasma proteins (50µl) were mixed with 1.5µl ZnAF-2, and then fluorescence (Ex 485nm/Em 528nm) was measured at baseline and following addition of 3.5µl aliquots of Zn²⁺ that increased total Zn²⁺ content in the sample by an increment of 8µM. Shown in Figure 9 are summaries of studies in controls (n=6) and animals subjected to mild surgical stress (n=6). Labile zinc signal is shown both before and after addition of ZnCl₂.

Figure 9.

Measurements of buffering capacity for Zn²⁺ within high- and low-molecular weight fractions of plasma proteins taken from control animals and animals undergoing mild surgical stress. Buffering capacity was assessed by mixing samples with ZnAF2 and measuring fluorescence before and after addition of a standard amount (8µM) of ZnCl₂. Responses of higher magnitude to the addition of exogenous Zn²⁺ indicate lower buffering capacity. Panels A and B provide information about buffering for samples (high-molecular weight concentrates and low-molecular weight filtrates) from control animals, while Panels C and D provide information for samples from animals undergoing mild surgical stress. A: in control animals subjected only to anesthesia, a very high level of binding capacity is observed within the high molecular weight fraction, demonstrated as nearly identical levels of labile zinc with (●) or without (○) added zinc. B: in control animals, a transient increase in binding capacity of the lower molecular weight fraction is observed and restored at 24 hrs. C: following anesthesia and mild surgical stress, significant decreases in binding capacity in the high molecular weigh fraction are observed and are not fully restored at 24 hours. D: in the stressed animals, the rapid responsiveness is lost that was observed under control conditions (Panel B). Data points are shown as mean ± SEM. * = p < 0.05 vs. Control, N = 6.

A first observation is that there is considerable affinity for Zn²⁺ in both the high molecular weight (concentrate or C) and low molecular weight (Filtrate or F) fractions. In the C fractions from control animals (Figure 9A), ZnAF-2 fluorescence was not altered following addition of 8 µM exogenous Zn²⁺, indicating a very high capacity for binding of Zn²⁺. Fluorescence in the F fraction (Figure 9B) was also much attenuated with addition of 8µM Zn²⁺, a level that would be expected to saturate the reporter. A second observation is that in periods B and C there was variability in the affinity for Zn²⁺ in the F fraction in control animals (Figure 9B), restored to baseline levels (period A) by the time measurements were obtained in period D.

Of principal note were the responses in samples taken from animals subjected to stress. As shown in Figure 9C, even mild surgical stress led to significant impairment of binding in the C fraction, in all periods following induction of stress. Conversely, stress seemed to eliminate the rapid responsiveness observed in control animals and, if anything, to enhance Zn²⁺ binding in the F fraction. Taken together, these observations are consistent with our hypothesis that even moderate surgical stress leads to significant redistribution of Zn²⁺ from the larger- to smaller-sized proteins. In addition, these findings suggest that concentrations of labile Zn²⁺ in the circulation remain remarkably constant due, in part, to such redistribution and alterations in affinity of plasma proteins.

DISCUSSION

Prior reports of assays for free Zn²⁺ in extracellular fluids have focused on use of dialysates (27, 32). In an early report (25) concentrations of Zn²⁺ were determined in equine plasma, utilizing an enzymatic bioassay, which required both time for equilibration, larger volumes and manipulation of conditions to exclude interference by Mg²⁺. In a more recent reports (34, 35), a fluorometric assay based on an early generation reporter for Zn²⁺, zinquin, was used to measure labile Zn²⁺ in the micromolar range in samples of plasma and other fluids, including cell-conditioned media. A key contribution of that report was the recognition that “lability” of the metal divalent cation is defined in large part by the selectivity, affinity and concentration of the reporter itself. Such studies were conducted on samples obtained from experimental subjects that were not acutely stressed. To our knowledge, measurements of labile Zn²⁺ in plasma samples have not been reported in patients who are acutely ill or animals that are under experimental stress.

In plasma, background fluorescence with excitation in the green range is not ideal, but manageable for semi-quantitative measurements. Optimal conditions for assay include immediate processing and use of minimal volume additions (<3%). In addition the reporter (~30nM) should be used in final concentrations that minimize, respectively, dilution of sample and significant chelation by the reporter itself. In the studies reported here, utilization of ZnAF-2 was guided by its selectivity for Zn²⁺ and high fluorescence yield, as well as the preservation of its dynamic range even in a complex fluid such as plasma.

If the reporter maintains its dissociation constant (K_d) for Zn²⁺ in a complex fluid such as plasma, then it is possible to utilize ZnAF-2 fluorescence to calculate approximate concentrations of free Zn²⁺. Utilizing the K_d (5nM) obtained in Ringer’s solutions, which accords well with published values, and the data obtained in Figure 1B, we may infer a free concentration of Zn²⁺ of 1nM to 3nM, under baseline conditions (see arrow in Figure 1A). These calculations agree reasonably with predictions and experimental observations suggesting that unbound Zn²⁺ levels in plasma are likely to be in the nanomolar range (26, 27, 28).

An important caveat of such calculations is that changes from baseline fluorescence levels are not linearly proportional to the changes in concentrations of labile Zn²⁺. In the middle of the dynamic range of the dye (about 5nM), they may be so, but not toward the margins of the response curve. Thus, when baseline concentrations rise or fall by 40%, as shown in Figures 2 and 5, there may be corresponding changes in actual concentration by a factor of 2 or 3. Such disturbances may seem small, because they occur in the nanomolar range. It must be remembered, however, that such changes in this range can initiate or terminate physiologic or pathologic processes, including initiation of pathways of apoptosis (36, 37) and degradation of extracellular matrix (38). The small scales on which such changes may occur should not obscure their potential importance in regulating activities of circulating cells and peripheral tissues. Nevertheless, the feasibility of fully quantitative measurements of labile Zn²⁺ in the circulation awaits development of dyes that do not require laborious corrections and normalizations against background fluorescence.

With this caveat, our studies with infusions of zinc with or without the nonspecific chelator, sodium citrate, confirm responsiveness of ZnAF-2 to changes in labile zinc in samples taken from the circulation. With infusions at 0.4 mg/kg (0.12mg for a 300g rat = 0.12mg/66mg/mmol=1.8uM), and given an initial volume of distribution limited to plasma (~0.2L/kg or 6ml for a 300g rat), we would estimate initial rise in total plasma Zn²⁺ content of 300µM/L. Such expected increases in initial plasma Zn²⁺ content would be detected at the upper limit of the colorimetric assay for total Zn²⁺, as was observed (Figure 2A). The observed rise in ZnAF-2 fluorescence accords well with that expected from an exogenous addition of Zn²⁺ to the highly buffered plasma (Figure 1B). Moreover, we show that this reporting system can be used to monitor changes in plasma Zn²⁺ when high capacity chelators, such as citrate, are simultaneously introduced into the circulation.

Our studies also demonstrate that ZnAF-2 can be used to interrogate the affinity of plasma, and its different protein fractions, for free Zn²⁺. Using ZnAF-2, we confirm that normal plasma has marked binding capacity for zinc, and that the binding capacity largely—but not exclusively—reflects binding by large molecular weight macromolecules (Figure 9) such as albumin and alpha-2 macroglobulin (18). We also provide evidence for binding in lower molecular weight (<10kD) fractions, which becomes proportionately more important during acute stress. Our studies indicate that changes in affinity can be monitored in large and small molecular weight fractions in samples taken directly from the circulation.

With characterization of the assay and its limitations, we have been able to investigate labile zinc distribution in a model of mild surgical stress. The model itself involves dissection, cannulation, and ligation of the femoral vessels in the rat hind limb. This degree of surgical stress has been characterized in previous studies as mild and is not associated with elevation in circulating markers of injury (39). In our model we show that plasma albumin, and acute phase reactant, is unchanged throughout the experiment. This finding confirms that alterations in total and free levels of circulating Zn2+ are not easily attributed to changes in levels of important binding proteins, but also confirming the mild nature of the surgical stress in this model. In this study, we find that even mild surgical stress leads to decreases in circulating Zn²⁺ overall and causes alterations in distribution of Zn²⁺ among different fractions of plasma proteins. The decline in total zinc levels was not attributable to concomitant decline in plasma albumin concentration, which did not change significantly in 24 hours following induction of stress. The subsequent studies using plasma fractions indicate that changes in zinc affinity of plasma proteins is the likely cause of the overall decline in total Zn²⁺ levels, and may also function to regulate concentrations of labile Zn²⁺ in the plasma that are observed throughout the study. The results of our experiments extend and clarify very early clinical reports that zinc in the plasma is unloaded to peripheral tissues during some forms of acute stress (40).

Our studies suggest that Zn²⁺ content of plasma can shift between high affinity, high capacity pools in high molecular-weight fractions to low affinity or labile pools during surgical stress. Much like the ionized calcium pool in acute calcium deficient conditions (41, 42), this redistribution of Zn²⁺ may occur so that it can more easily be transferred to meet demand in peripheral tissues. Our observations also emphasize that anesthesia and moderate surgical stresses can elicit subtle, acute responses in delivery and clearance of metal ions from the circulation. Modification of plasma albumin metal ion(Cobalt) binding affinity has been shown to occur in response to organ ischemia (43, 44). Our model characterizes the responses of both the total circulating zinc pool and the small labile fraction in acute stress; our results indicate that total zinc and labile zinc both decrease in mild stress, and that zinc binding increases in the lower molecular weight plasma fraction. Our results and other reports (43, 44) suggests that metal ion binding has a dynamic role in the metabolic response to stress, and can be studied independently of other markers. In addition, chronic zinc deficiency and zinc responsive conditions such as chronic steroid use may be studied in terms of labile zinc affinity.

Acute changes in zinc availability have not been recognized until now as having clinical significance. Chronic Zn²⁺ deficiency is well recognized as a risk factor for infection and poor healing. Increasingly, however, it has become clear that demand of peripheral tissues and parenchymal cells for Zn²⁺ may occur in response to endocrine stimulation (45), normal physiologic activity (46), as well as oxidative stress (47,48) and systemic sepsis (24). Rapidly deployable assays for free Zn²⁺ in extracellular fluids such as the one described here are crucial for development of studies to understand how such acute demand for Zn²⁺ is satisfied and to identify experimental conditions and clinical circumstances in which the capacity of the circulation to deliver Zn²⁺ is insufficient to meet the demand. Clinical conditions previously identified as zinc deficiency may be need to re-defined in the context of zinc utilization and distribution.

Abbreviations

Zn2+

ionized divalent zinc

EGTA

Ethylene glycine tetra acetate

K_d

Dissociation constant