Abstract
Objective:
The locations of sodium (Na) storage and its exchange mechanisms in the body are not well known. Understanding tissue Na storage and exchange is important for understanding the impact of Na intake, absorption, and retention on human health, especially on the risk of developing chronic diseases. The purpose of this study was to investigate the application of a deuterium–deuterium (DD) neutron generator-based IVNAA system in Na nutrition studies.
Approach:
The right legs of two live pigs, one on a low Na diet and one on a high Na diet, both for 14 d, were irradiated inside a customized irradiation cave for 10 min (45 mSv dose to the leg) and then measured with a 100% efficient high purity germanium detector (HPGe). The spectra were analyzed to obtain the net Na counts at different time points. Bone Na concentrations were calculated using the calibration created with Na bone phantoms.
Main results:
The results show that the difference in bone Na to calcium between the pigs on high versus low Na diets was 466 ± 137 ppm. The estimated bone Na to calcium concentrations were 1166 ± 80 and 1631 ± 111 ppm for low and high Na diet pigs, respectively. Analysis also shows rapid exchange of Na in the leg during the first 2 h measurements, while the exchange was minimal at the second and third 2 h measurements, taken 7 and 21 h post irradiation. The exchange decay time of Na in the leg was 51 min for the first measurement, and there was no significant change of Na activities between 2–21 h.
Significance:
With these results, we conclude there is a non or low exchangeable compartment (likely to be bone) for Na storage and that DD neutron generator-based IVNAA is a useful method for determining tissue Na distribution in nutrition studies.
Keywords: sodium, bone sodium, in vivo neutron activation analysis (IVNAA), deuterium–deuterium (DD) generator
1. Introduction
The purpose of this study was to investigate the application of a customized compact deuterium–deuterium (DD) generator for in vivo neutron activation analysis (IVNAA) in sodium (Na) nutrition studies. The system was previously optimized and calibrated for human hand irradiation (Coyne et al 2018) and was adapted for use with live pigs. The two primary objectives of the study were to investigate the use of the system for detecting changes in bone Na from dietary intervention and to investigate the influence of dietary sodium on tissue Na storage and exchange in the body.
Na is an essential element in the body. However, too much Na in the exchangeable fluid can result in high blood pressure which in turn can increase incidence of congestive heart failure or diseases of the liver and kidney (Titze et al 2014). Na Dietary Reference Intakes are typically exceeded by individuals in the US. The storage site of Na in tissue is likely to influence its ability to influence blood pressure, depending on whether fluid retention is affected. Thus, there is a pressing need to better understand how Na is metabolized and stored in the body and the influence of perturbing factors such as dietary Na intake on tissue Na storage and, ultimately, disease progression.
Typically, Na is measured within biologic samples such as urine. However, the validity of current methods used to estimate Na intake from free-living individuals is under question since new data show that urinary Na excretion does not mirror Na intake on a day-to-day basis (Weaver et al 2016). This may be explained by the deposit of Na in bodily tissues (Machnik et al 2009). Indeed, bone is a storage organ for many metals, including Na (Massey and Whiting 1996). Na is primarily stored in the matrix of cortical bone. The effect of bone Na stores on urinary Na excretion and estimates of Na intake, as well as the effect of these stores on human health, remain to be elucidated. In order to do this, a non-invasive method for quantifying bone sodium content in vivo is needed. In this study, a novel IVNAA technology was applied to noninvasively quantify Na in pig bone. Because the produced radionuclide Na-24 has a physical half-life of 15 h, its exchange between the pig leg and the rest of the pig body was also studied.
2. Materials and methods
2.1. In vivo NAA (IVNAA) system setup and measurement of Na
The neutron source for irradiation was a DD109M neutron generator (Adelphi Technology, Inc., Redwood, CA). The DD109M is a customized compact deuterium–deuterium generator that produces a close-to-isotropic source of 2.45 MeV neutrons. The flux can be adjusted with current and voltage to produce a flux of up to 5 × 109 n s−1. The generator has a modifiable moderator/reflector/shielding assembly previously optimized for human hand irradiation (Coyne et al 2018). This study used a larger irradiation cave with adjusted reflector/shielding assembly to accommodate a young pig.
With neutron activation of Na, low energy neutrons interact with Na-23 and produce Na-24 with two major γ-lines with energies of 1368 keV and 2754 keV. The 100% abundance of Na-24, along with the high branch ratios of 90% for 1368 keV and 100% for 2754 keV, accommodate high sensitivity in the detection of activated sodium. Similarly with calcium (Ca), Ca-48 is activated to produce Ca-49 with a distinct γ-line of energy 3083 keV. Since the activated Na and Ca counts are from the same neutron flux, the normalization of Na signal to Ca signal is expected to correct for the variation of several factors including neutron flux, leg attenuation, and counting geometry factors.
The net counts from Na and Ca characteristic γ-rays are calculated by an in-house spectral analysis program. The γ-ray detector used for this study was a 100% high efficiency ORTEC GEM100P4-95 HPGe (AMETEK, Oak Ridge, TN). The software used to collect the γ-ray spectra is MAESTRO by ORTEC (AMETEK, Oak Ridge, TN).
2.2. Pig characteristics and dietary intervention
All study personnel who interacted with the pigs completed the appropriate trainings, and all study procedures were approved by the Animal Care and Use Committee at Purdue University. This study was completed on two male domestic commercial pigs. Pigs were acquired shortly after weaning and housed together in the large animal facility at Purdue University. Pigs acclimated to the facility and gradually transitioned to the study diet for 4 d. On day 5 pigs were fully transitioned to diet. Pig 1 was fed a low Na diet (0.2% Na w/w) and pig 2 was fed a high Na diet (0.6% Na w/w) (Research Diets, Inc). Both diets were a pelleted casein-based growing swine feed, and Na was provided in the form of NaCl. Both pigs consumed their diet for a full 14 d before IVNAA measurements were conducted. At the time of the IVNAA measurements, the low and high Na pig weighed 16.8 kg and 17 kg respectively and both had a leg circumference of 15 cm at the irradiation site.
2.3. IVNAA for Na measurement in pig leg and leg bone
The right posterior leg of a live pig was irradiated for 10 min with the center of irradiation aimed as shown in figure 1. During irradiation and subsequent detections, pigs were anesthetized to prevent movement and intubated to maintain a clear airway while under anesthetic. Vital signs were monitored every 10 min to ensure the safety of the animals.
Figure 1.
Location of irradiation on pig leg.
The irradiation cave was modified for this study in order to comfortably accommodate the pig, as shown in figure 2. The non irradiated left leg was shielded and an electronic dosimeter was placed to record the given dose. Both pigs received an extremity dose of 45 mSv to the right hind leg, which agrees with the anticipated dose from simulation and calibration (Coyne et al 2018). Accounting for a tissue weighting factor of 0.01 for the leg skin and bone surfaces, and with the irradiated area estimated to be 1.25% of the total body weight, the effective dose to each pig was 5.63 μSv.
Figure 2.
Modified irradiation cave for pig study.
Following irradiation, the pigs were moved to the detector as shown in figure 3, where consistent leg placement was visually confirmed for each measurement.
Figure 3.
Detection of irradiated pig leg.
The pig legs were measured with the high purity germanium detector (HPGe) for up to 2 h (with 5 min counting intervals) at decay times of 4 min, 7.5 h, and 21.5 h. The pigs were anesthetized during the measurements, but were awake and able to move between measurements. The anesthesia process contributed to slight differences in detection start and end times between pigs. Following the final measurement, the pigs were euthanized, and both hind legs were removed and dissected.
2.4. In vitro NAA for Na measurement in pig leg and leg bone.
All four hind legs were irradiated and detected following pig dissection. The legs were first irradiated as a whole piece and then dissected into muscle, skin and bone groups.
3. Results
3.1. Na concentration
3.1.1. In vivo spectrum and first 2 h measurement
Figure 4 below shows an in vivo spectrum of the low Na diet pig during the first 60 min of detection.
Figure 4.
Low Na pig in vivo spectrum.
Spectral analysis was performed to determine the net Na and Ca counts for each pig during the first 60 min of measurement. The ratio of Na to Ca counts is used in order to normalize the Na counts to bone Ca for each pig, as shown in figure 5. Since Ca has a short half-life of 8.7 min, the total Ca counts during the first 10 min of detection were used to normalize each Na count interval. All Na counts were decay corrected to represent Na counts at time 0, immediately following irradiation. The decay corrected intervals show higher Na counts per Ca counts for the high Na diet pig, which is also observed with the 2754 keV Na counts.
Figure 5.
First measurement, low Na versus high Na diet pig Na/Ca counts at 1368 keV.
The linear trend lines from the previously obtained bone Na concentration calibration curves (Coyne et al 2018) were used to approximate Na concentration of the pig leg. Using the first 1 h in vivo sum of Na and Ca counts, the low Na pig is estimated to have a Na/Ca concentration of 2441 ± 186 ppm while the high Na pig is estimated to have a Na/Ca concentration of 2812 ± 232 ppm. This results in a concentration difference of 372 ± 297 ppm (p = 0.22).
3.1.2. Comparison among low and high Na diet pig at second 2 h measurement and bone Na
The initial measurement reflects Na signal from the entire leg, including highly exchangeable blood. It is hypothesized that bone Na concentration is better reflected in later measurements, when the activated Na in blood has left the leg and the stable Na signal reflects Na in slowly exchanging compartments, like bone. Figure 6 shows the comparison in Na/Ca ratio during the second 2 h measurement, 7 h post irradiation. The figure is corrected for radioactive decay and counting time and shows the high Na diet pig has a higher normalized ratio than the low Na diet pig. A similar trend is seen during the third 2 h measurement, 14 h post irradiation, and both energies of Na show the same pattern.
Figure 6.
Second measurement, low versus high Na diet pig Na/Ca counts at 2754 keV.
The linear trend lines from the previously obtained bone Na/Ca calibration curves (Coyne et al 2018) are used to approximate Na/Ca concentration of the pigs. Using the calibration curve adjusted for 7 h of decay and the in vivo measurements averaged during the second measurement, the low Na diet pig is estimated to have a Na/Ca bone concentration of 1165 ± 80 ppm while the high Na diet pig is estimated to have a Na/Ca concentration of 1631 ± 111 ppm. This results in a concentration difference of 466 ± 137 ppm (p = 0.041). Since the amount of activated Na is relatively stable during the second measurement, these concentrations most likely reflect bone content. These results exhibit a clear separation in bone Na concentration over a short dietary intervention time, for the two pigs studied.
3.1.3. In vitro comparison
Following the completion of the in vivo study, the removed pig legs were first irradiated whole as in vitro samples, with the same irradiation protocol as the in vivo study. Using the bone Na phantom calibration, the low Na diet pig had an in vitro Na concentration of 5547 ± 331 ppm and the high diet Na pig had a concentration of 6836 ± 433 ppm, resulting in a difference of 1288 ± 545 ppm. However, the calibration was not intended for in vitro analysis, so these estimated concentrations are provided to further demonstrate the difference in Na concentration between the low and high Na diet pigs, as seen in the in vivo study. The in vitro concentrations are higher due to the lack of circulation and exchange of Na in the leg, and shows the significant contribution of Na from the fast exchange compartment that decreases during in vivo measurements. Additionally, the in vitro concentration is higher due to the difference in irradiation and detection geometry between a live pig and a dissected leg. Despite the differences, these data can be used for designing future in vivo studies.
In order to determine where Na is stored in the body, the legs were dissected and irradiated under the same conditions as the in vivo study. The results show Na is present in all categories of dissected leg- bone, muscle, skin and connective tissue. Ca was only found in bone. As shown in figure 7 for 2754 keV, there is significant storage of Na in the short bones, which are the primary target of irradiation. These results were mirrored at 1368 keV.
Figure 7.
In vitro net Na counts at 2754 keV.
The Na counts were normalized to total weight of disected pieces to yield Na per gram as shown in figure 8.
Figure 8.
Na counts at 2754 keV normalized to weight.
The two hour bone Na counts normalized to 1 h Ca counts are shown for all 4 legs in figure 9 for 2754 keV. Accounting for error propagation, the in vitro bone results do not separate as distinctly as for in vivo results due to dietary Na levels. Variance was performed on the in vitro Na/g data to determine if there is a significant difference between the groups studied. These results are shown in table 1. As expected, there is not a significant difference between the data sets from the two energies of Na, which matches the trend shown in the in vivo data. There is also not a difference between Na counts measured in the right hind leg compared to the left. The difference in Na/g between the high and low Na diet pig is marginally significant, based on the in vitro data. When looking at just the results from 1368 keV, which has a higher detection efficiency, there is a significant difference (p = 0.028, n = 12) between the dissected samples from the high and low Na diet pigs. More pigs are needed to confirm these results, but the in vitro data confirms the in vivo conclusion for the two pigs studied.
Figure 9.
Na/Ca ratio at 2754 keV.
Table 1.
In vitro variance with n = 24.
| Comparison | P-value | Significant? |
|---|---|---|
| Energies (1368 and 2754 keV) | 0.85 | No |
| Legs (left and right) | 0.49 | No |
| Pigs (high and low Na diet) | 0.06 | Marginally |
3.2. Na exchange and storage
In order to examine the exchange of Na in the leg bone, radioactive decay is corrected with the use of saturation, decay and counting factors specific to Na-24 (Coyne et al 2018). In this study, the time of irradiation was 10 min, while the decay and count time varied throughout the measurements. The corrected activity over time is plotted in order to find the exchange half-life, te, from the equation m = ln2/te, where m is the slope from the fitted equation γ = Ae−mx. In order to determine Na exchange times, decay corrected interval Na counts were normalized to a one hour Ca count.
For the first measurement post irradiation, the radioactive decay corrected exchange of Na in the low Na diet pig is shown in figure 10. The low Na diet pig only completed 80 min of the first 2 h detection session due to issues with anesthesia. The radioactive decay corrected exchange of Na/Ca during the first measurement in the high Na diet pig is shown in figure 11.
Figure 10.
Low Na diet pig fast Na exchange.
Figure 11.
High Na diet pig fast Na exchange.
To investigate slow exchange of Na, the 2 h measurements taken at 7 and 21 h post irradiation are used. For the second and third measurements, the radioactive decay corrected exchange of Na/Ca in the low Na diet pig is shown in figure 12.
Figure 12.
Low Na diet pig slow exchange of Na.
Figure 13 shows the same measurements in the high Na diet pig. For both pigs, the last measurement point from the first 2 h measurement is relatively stable with the second and third hour measurements. This indicates that the fast exchange half-life can be roughly approximated from the first 90 min of measurement for both pigs. Using the 80 min measurement, the low Na diet pig has an approximate fast exchange half-life of 45 min. The fast exchange half-life of the high Na diet pig based on the first 90 min of measurement is estimated to be 56 min. This results in an average Na fast exchange half-life of 51 min for this study. Analysis shows rapid exchange of Na in the leg during the first 2 h measurements, while the exchange was minimal at the second and third 2 h measurements, taken 7 and 21 h post irradiation. There was no significant change of Na in the leg between 2–21 h.
Figure 13.
High Na diet pig slow exchange of Na.
4. Discussion and conclusion
The IVNAA system is able to non-invasively detect differences in tissue Na in pigs consuming low versus high Na diets for a relatively short period of time (14 d). The in vivo measurement showed a significantly higher Na/Ca ratio at the second 2 h measurements. The difference is also observable at the first and third 2 h measurements. The in vitro analysis shows Na stored in all measured tissues including bone, muscle, skin and soft tissue, as well as significantly higher concentrations of Na/g in bone, soft tissue, and skin. The in vitro results confirm a significant difference in Na between the high and low Na diet pigs studied.
The estimated fast exchange half-life of 51 min agrees with a past IVNAA study performed with Californium-252 in human hand that found a fast exchange half-life of 1 h (Cohen-Boulakia et al 1981). The Cf-252 study found exchange half-lives of 79 h in normal bone and 35 h in bone with low Ca concentration. The direct relationship in bone Na and Ca concentration suggests that the slowly exchangeable compartment behaves the same as bone Ca, although more analysis is needed to show the mechanisms of Na deposition and exchange over time. Further, the ratio of Na to Ca is consistent throughout the skeleton, indicating the slow exchange compartment may be independent of the fast exchange compartments. (Cohen-Boulakia et al 1981)
The fast exchange compartments are hypothesized to be soft tissue, muscle, and/or skin. Skin has been observed as an osmotically inactive storage site for excess salt from dietary sources (Titze et al 2003). Na MRI studies show Na storage in the epidermis with a hypothesized countercurrent system for electrolyte exchange (Hofmeister et al 2015). During the next pig study, the kinetics of Na exchange and storage in soft tissue will be studied with a Na coil MRI and direct chemical Na analysis of tissues to complement the IVNAA bone measurements.
These findings suggest that dietary Na intake affects tissue Na storage, and that a higher Na diet increases these tissue Na stores. Understanding the storage and exchange of Na will have important implications for human health, especially for chronic diseases such as cardiovascular disease and osteoporosis. With the in vivo and in vitro pig study results, we conclude there is a non or low exchangeable compartment (likely to be bone) for Na storage and that the DD neutron generator-based IVNAA is a useful method in Na nutrition studies.
There are important limitations to this study, however. More pigs are needed to verify our results with true statistical significance. More in vivo measurements with longer count times would help clarify the exchange time and reduce radiation counting errors. Additionally, 14 d may not be enough time for excess Na to deposit in bone. Future studies will be conducted to address these limitations and further investigate the potential of the system with refined methodology. A larger sample size will be used to clarify the variation in response to dietary Na. A precision study is planned to determine the reproducibility of the measurement. Baseline IVNAA measurements will be added before the dietary intervention starts to better observe the effects of the diet on Na concentration and exchange. The addition of MRI measurements will also help understand the storage and exchange of Na in soft tissues, but not bone.
The system is also anticipated for use in human studies with planned applications in dietary Na interventions. The use of IVNAA will help quantify the long-term effects of dietary change on bone Na and the relation to Ca retention, since research shows bone Ca and Na are deposited and stored similarly. IVNAA has the potential to be instrumental in understanding Na mechanisms in the body and in the prevention of Na related diseases.
Acknowledgment
This research was supported by the NRC Nuclear Engineering and Health Physics Fellowship to M Coyne, Purdue Ross Fellowship to M Coyne, and Indiana CTSI Project Development Team (PDT) pilot grant to L H Nie. We thank Dr Ellen Wells for advice on performing statistical analysis.
References
- Cohen-Boulakia F, Maziere B and Comar D 1981. Application of in vivo activation analysis to the compartmental study of sodium in the hand Phys. Med. Biol 26 857. [DOI] [PubMed] [Google Scholar]
- Coyne MD, Neumann C, Zhang X, Byrne P, Liu Y, Weaver CM and Nie LH 2018. Compact DD generator-based in vivo neutron activation analysis (IVNAA) system to determine sodium concentrations in human bone Physiol. Meas 39 055004. [DOI] [PubMed] [Google Scholar]
- Hofmeister LH, Perisic S and Titze J 2015. Tissue sodium storage: evidence for kidney-like extrarenal countercurrent systems? Pflugers Arch. 467 551–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Machnik A et al. 2009. Macrophages regulate salt-dependent volume and blood pressure by a vascular endothelial growth factor-C-dependent buffering mechanism Nat. Med 15 545–52 [DOI] [PubMed] [Google Scholar]
- Massey LK and Whiting SJ 1996. Dietary salt, urinary calcium, and bone loss J. Bone Miner. Res 11 731–6 [DOI] [PubMed] [Google Scholar]
- Titze J, Lang R, Ilies C, Schwind KH, Kirsch KA, Dietsch P, Luft FC and Hilgers KF 2003. Osmotically inactive skin Na+ storage in rats Am. J. Physiol. Renal Physiol 285 F1108–17 [DOI] [PubMed] [Google Scholar]
- Titze J, Müller DN and Luft FC 2014. Taking another ‘look’ at sodium Can. J. Cardiol 30 473–5 [DOI] [PubMed] [Google Scholar]
- Weaver CM, Martin BR, McCabe GP, McCabe LD, Woodward M, Anderson CA and Appel LJ 2016. Individual variation in urinary sodium excretion among adolescent girls on a fixed intake J. Hypertension 34 1290–7 [DOI] [PubMed] [Google Scholar]