Temporal Exposure of Dermal Fibroblasts to Silver Diamine Fluoride and Fluorine NMR Quantitation

INTRODUCTION

Silver diamine fluoride (SDF) has been advocated as a caries arresting material for Early Childhood Caries [1]. Clinical application of SDF has been reported to arrest the caries process for a period of up to 6 months thereby requiring subsequent re-applications [2]. This may pose a problem with non-compliant families that delay or ignore follow up visits. In addition, a systematic review of the clinical trials reveals that many trials did not include positive controls, often comparing SDF to either no treatment or just a placebo [3]. Normally, clinical trials would include both negative and positive controls. Interestingly, a one-time application of SDF did not significantly prevent dental caries [4]. Therefore, due to the requirement for multiple treatments, there is a need for SDF cytotoxicity studies over time.

Cytotoxicity tests on current concentrations of SDF to soft tissue should be thoroughly assessed and analyzed at selected time intervals. Also, the potential long-term effect of SDF on mitochondria should be investigated. A study by Lou et al on the influence of chronic fluorosis on mitochondrial dynamics morphology and distribution in cortical neurons of the rat brain indicated that chronic fluorosis induces abnormal mitochondrial dynamics, which might in turn result in a high level of oxidative stress [5]. In addition, the effects of fluoride ion in dental medicaments on cell vitality, specifically the mitochondria should be investigated. NMR analysis of fluoride ion levels may be a technique useful in determining if fluoride ions are anti-mitochondrial and should be further investigated as mitochondrial health is linked to aging and cancer. Recently published studies on the possible effects of chronic over exposure to fluoride have created increased interest in the dental products using fluoride. Twenty-one of twenty-three epidemiology studies have reported a possible negative effect on cognitive abilities or intelligence due to chronic fluoride exposure [6]. Animal studies have presented plausible biochemical mechanisms for the neurologic effects [7]. The effect of fluoride on the brain microglia has also been established [8]. Oddly enough, the anti-microbial nature of the fluoride ion that is partially responsible for dental benefits, may be the reason for the anti-mitochondrial effect. For example, anti-bacterial antibiotics are also anti-mitochondrial [9]. The use of non-dental personnel in an un-controlled environment with lack of proper suctioning and lighting may contribute to un-necessary exposure to any cytotoxic effects. In addition, the risk benefit ratio should always be considered.

Fluoride varnish was once touted as an effective preventative for young children [10]. Unfortunately, a systematic review of Fluoride varnish clinical trials revealed a discrepancy between the much-heralded preventative predicted effectiveness and it’s actual performance [11]. Indeed, of nine clinical trials, only 3 showed any positive benefit in little children, although fluoride varnish may be recommended for older patients. Other interventions in these three studies may have been responsible for the positive result. Additionally, the release of fluoride into the saliva by fluoride varnishes does NOT correlate with the absorption of fluoride into the enamel [12]. Possibly, the release of fluoride into the saliva only gives a systemic dosing to the recipient and may actually be associated with more potential side effects and little enamel protection. Perhaps the “Old Standard” of Acidulated Phosphate Fluoride foam/gel treatment for younger children under the age of 6 should still be utilized as a dental preventative [13]

Fluoride induces cell stress, including endoplasmic reticulum stress and oxidative stress, which leads to impairment of ameloblasts responsible for dental enamel formation. Recently it was reported that fluoride activates SIRT1 and autophagy as an adaptive response to protect cells from stress [14]. Interestingly, fluoride dentifrice was the main source of fluoride intake by the children evaluated. However, the fluoride concentration in food items also significantly contributed to the daily ingestion by 2-6-year-old children [15]. Most of children in the study were exposed to a daily fluoride intake above the suggested threshold for dental fluorosis. The dentifrice alone is responsible for an average of 81.5% of the daily fluoride intake, while among the constituents of the diet, water and milk are the most important contributors [16]. Children ingested around 60% of the dentifrice loaded onto the brush, but no significant differences were seen among the groups (p > 0.05). Mean daily fluoride intake from dentifrice for the three groups (G-A, G-B and G-C) was 0.022(a), 0.032(a) and 0.061(b) mg F/kg body weight, respectively (p < 0.01). There was a strong positive correlation (r = 0.86, p < 0.0001) between the amount of dentifrice used and the amount of fluoride ingested during toothbrushing. The results indicate the need for instructing children’s parents and care givers to use a small amount of dentifrice (< 0.3 g) to avoid excessive ingestion of fluoride [17].

Children with low height-for-age were more likely to have dental fluorosis in the TFI (Thylstrup-Fejerskov Index) categories that affect the entire tooth surface. The results suggest that subpopulations with chronic undernutrition are more susceptible to dental fluorosis [18]. If future studies are done, this preliminary information should help in designing SDF assays going forward. For now, it is fair to assume that the actual SDF cytotoxic endpoints are lower than what has been obtained from our DF culture studies. Among human populations, fluoride-related health issues span from mild effects on teeth and bones to severe kidney problems, neurotoxicity, and even cancer [19, 20, 21, 22, 23]. Nowadays, fluoride toxicity is gaining much attention in the fluoride endemic areas due to its adverse impact on children’s cognitive function, since children are more susceptible to fluoride toxicity as compared to adults. The primary mechanism behind the neurotoxicity of fluoride is still not clearly understood. However, fluoride reportedly disrupts the biochemical mechanism and thus alters the normal functioning of the brain. Several human studies are suggestive of lowered IQs in children exposed to high fluoride during their childhood [24]. The conclusive findings of previous published research suggest that the fluoride concentration in drinking water tends to grossly affect the children’s IQ level [25]. However, proof is limited, because most of the studies are concentrated in a few countries and have limitations and data gaps [26]. The grave potential implications of fluoride toxicity demand extensive and meticulously designed research to bridge the data gap in developing countries. Furthermore, spanning research work across all the fluoride endemic areas of the world could generate a larger database, helping resolve the issue on a large scale [27].

Unfortunately, the mothers’ report estimating the amount of toothpaste on the toothbrush using demonstrative figures was not a reliable method to judge the inadvertent ingestion of fluoride from dentifrices [28]. Fluoride ingestion is probably much greater than estimated, which may cause one to speculate as to the greater over-all effect. Some parents may be less concerned because they utilize a “children’s” toothpaste. However, the findings suggest that the ingested dose should be calculated based on TSF (Total Systemic Fluoride). The dose of TSF ingested by children is similar whether adult or children’s toothpaste is used [29]. The majority of dentifrices used in that study contained F (96%) and in 84% of them, Total Fluoride concentration was representative of that declared by the manufacturers. In the F-toothpastes, 78% showed TSF concentration ≥ 1000 ppm, varying from 422.3 to 1432.3 ppm F (mean ± SD of 1017.6 ± 239.4) [30]. So most children in that study were exposed to a high fluoride concentration [29]. However, a high concentration is necessary to prevent decay according to published results if only fluoride is considered. “There are benefits of using fluoride toothpaste at certain strengths to prevent tooth decay when compared with non‐fluoride toothpaste. The stronger the fluoride concentration, the more decay is prevented. For many of the comparisons of different strengths of toothpaste, the findings are uncertain and could be challenged by further research. The choice of fluoride toothpaste for young children should be balanced against the risk of fluorosis,” reported a Cochrane database article [31]. This guideline does not consider potential side effects or the reported neurologic toxicity. Frequent ingestion of low but excessive quantities of fluoride during the period of tooth formation can certainly lead to dental fluorosis. That risk is well understood. But particular concern is warranted for the ingestion of fluoride-containing toothpastes by young children and the inappropriate use of dietary fluoride supplements in communities with sufficient fluoride already present in drinking water [32].

MATERIALS AND METHOD

Dermal Fibroblast (DF) cells were grown in 24 well cluster plates to confluency. For experimental procedures, test wells were fitted with 0.4 micron Transwell^TM inserts (depicted in Figure 1.). Each well of DF cells contained 0.9mL of DF culture medium. DF cells were challenged with 0.1 mL of SDF in sterile water in the Transwell insert chamber to achieve a final concentration of 0.03% SDF in 1 ml total volume. Cultures were reincubated for 30 minutes, 1, 2, 4 and 8 hours. SDF at levels above 0.008% precipitate in the presence of culture media forming an insoluble precipitate as determined in preliminary studies (Data not shown). Trans-wells were employed to capture precipitate but permit the soluble SDF portion to pass through the membrane. At the selected time points Transwell inserts were removed. SDF culture media was removed and saved for fluoride assay. Fresh culture media (1.0mL) was added to the harvested wells and allowed to re-incubate up to 8 hours. Harvested SDF culture media was centrifuged @ 15,000 x g for 10 minutes, harvested and stored at -70°C for fluoride assay. After 8 hours, all remaining Transwell inserts were removed, media was aspirated and saved while DF cells were fixed with 95% Ethanol for 10 minutes and then stained with methylene blue. All DF cell images were taken at 600X. For additional NMR studies SDF was prepared diluted in water, in DF media at a 10-fold concentration and in normal saline at the same concentrations used in the DF studies. All samples were subjected to centrifugation at 15,000 x g to remove any resulting precipitates and supernatants were harvested for the NMR (nuclear magnetic resonance) studies.

Figure 1: Diagram of a Transwell linsert – courtesy of Vandateb. com (Transwells- Corning Life Sciences One Riverfront Plaza, Corning, NY 1483 The lower compartment contains 0.9mL of DF culture media . 0.1mL of SDF at 0.3% is added to the upper chamber.

RESULTS

The temporal cell changes to SDF exposure are seen in Figures 2 through 7 below. Morphological changes (Initial changes from striated cells to rhomboid appearance followed by cell loss) are clearly seen within 1 hour exposure with complete loss of all cells by 8 hours.

Figure 2. Untreated Dermal Fibroblasts (DF).

Figure 3. Dermal Fibroblasts after 30 min. exposure to 0.03% SDF (600X magnification exposure to SDF (600X magnification)

Figure 4: Dermal Fibroblasts after 1 hour exposure

Figure 5: Dermal Fibroblasts after 2 hour exposure to 0.03% SDF (600x magnification) to 0.03% SDF (600X magnification)

Figure 6: Dermal Fibroblasts after 4 hour exposure

Figure 7: Dermal Fibroblasts after 8 hour exposure to 0.03% SDF (600X magnification) to 0.03% SDF (600X magnification)

NMR Fluorine quantitation in Dermal Fibroblast (DF) Culture system:

Silver Diamine Fluoride (SDF) used for dental applications is supplied as a 38% (2.36M) unbuffered aqueous preparation. Cytotoxicity studies in the SID lab have shown that SDF is toxic to dermal fibroblasts in unfiltered concentrations as low as 0.002% (0.116mM) after the 24 hour exposure to DF cells. When SDF is applied to the cells separated by a 0.4 micron Transwell^TM membrane the toxic endpoint is slightly higher (0.004%). The reason for using Transwell systems is that SDF precipitates instantly when exposed to proteinaceous complete culture media, normal saline (0.85%) and a host of other common chemical entities. The characteristic precipitation of SDF confounds our ability to measure the actual concentrations of SDF or its ionic fluoride component that target the DF cells in a cytotoxic manner. Typical environmental Fluoride assays require large volumes of liquid preparations to quantitate the ion and are impractical for small volume biological assay systems. For that reason, we turned to a more sensitive NMR based assay which was recommended by the Northwestern University Chemistry department and conducted in their IMSERC NMR facility. Our sample sizes available were sufficient to estimate the amount of fluoride generated post SDF precipitation that might be expected to affect DF cells.

Preliminary NMR Study: Determination of detectability of SDF/F- in DF Culture media.

The preliminary results suggest that fluoride can be detected at relatively low concentrations but the presence of DF media components may interfere to some extent on the quantitative detection of fluoride ion when binding occurs in minor DF media components. We re-examined a higher concentration of SDF in DF medium with an internal Trifluoroacetic acid (TFA) standard. Figure 11 shows the NMR results of 0.3% SDF on DF media post centrifugation. Triflouroacetic acid (TFA) is used a quantitation standard as described in the methods section. TFA standard peak is seen at -75.4 ppm as a single peak. Three SDF related fluoride peaks are seen at ppm positions -130.06, -143.33, and -157.85 in the presence of DF media. The multiple peaks are not unexpected for SDF because of the complex culture media matrix that potentially bind fluorine. The ppm frequency shifts are also due to the pH changes caused by TFA. The values for each peak relative to the ~10mM TFA peak are shown in the plot table. All of the peaks are in the μM range. The total fluorine recovered is the sum of the three peaks.

Figure 8. Generic Diagram of NMR Spectrometer ( From Barron, P. M. V. R. & A. R. (2020, June 9). NMR Spectroscopy–https://chem.libretexts.org/@go/page/55887)

Figure 9. Two standard qualitative NMR plots are shown to assess background fluoride in DF medium. The upper plot is 0.06% (3.72mM) SDF in DF media. A single small fluoride peak can be seen at -120ppm position (see arrow).The lower black line is a plot of DF media without any SDF showing no fluorine peak at –120 ppm or elsewhere in this expanded plot.

Figure 10. Expanded 0.06% peak. The broad peak is due to fluorine binding to media components which is further evaluated below.

Figure 11. TFA SDF Peaks

Figure 12: 0.03% SDF in H₂0

As stated, normal (0.85%) saline precipitates SDF in the same way as DF media without the additional binding entities. That said, we replicated our DF studies using normal saline as a precipitant using Transwells to capture precipitated SDF and measured Fluorine as described. We also looked at SDF in water at the 0.03% SDF level.

To determine the NMR position of SDF in an aqueous environment, we looked at a 0.03% (1.86mM) preparation of SDF in molecular grade water. As can be seen, a very well defined single peak at -119.4 is generated that is very narrow in the absence of other molecular entities. There was no visible precipitate post-centrifugation with this aqueous preparation. Estimated recovery ~0.2836mM (15.2%) relative to absolute NMR value.

As seen in DF media, the TFA changes the matrix pH and the frequency shifts the SDF peak position to -161.41. In this case some SDF precipitates in the presence of normal saline and some signal (~ one third) is lost compared to that in H₂O.

Figure 13 (below) is a composite of 6 samples normalized to TFA standards used to estimate SDF (F-) concentrations and to demonstrate the SDF fluorine ppm shifts that are related to the use of TFA as an internal standard compared to samples assayed without TFA Ultimately, the peak shifts do not significantly affect quantitation of fluorine which is supported by the assay linearity check in Figure 14 below. The spectra below are (from top to bottom) 0.004% SDF in saline, 0.008% SDF in saline, 0.015% SDF in saline, 0.03% SDF in saline with TFA, 0.03% SDF in H₂O, and 0.3%sdf in DF media with TFA.

Figure 13: NMR of 0.03% SDF in Normal saline with ~10mM TFA internal standard.

Figure 14. Linearity of Fluorine NMR Assay y for SDF in Saline:

Table 1 summarizes the SDF data and Figure 14 below plots the absolute quantitative NMR peak value vs. the nominal SDF input value. As can be seen, relative quantities of fluoride can be assessed with NMR. It is clear that the fluorine levels from SDF can be measured in the in-vitro system used in this study. The levels of SDF are small but clearly toxic to DF cells. But only represent a small fraction of fluorine from the nominal input SDF. We can also see that we can measure differences in fluorine in a dose dependent manner in a linear manner.

Table 1.NMR Data Summary

Nominal SDF input concentrations	Absolute NMR SDF values	Fluorine recovery (% SDF)	{SDF} normalized to TFA Standard	Fluorine Peak values (ppm)
	μM values		μM values
0.3% SDF in DF media	1173.4	6.3%	425.49	Peak 1: (-130.6) Peak 2(-143.33), Peak 3: (-157.85)
0.03% SDF in H2O	283.63	15.2 %	N/A	(-119.4) to (-120.65)
%SDF in 0.85% NaCl
0.03% (1860 µM)	190.35	10.2 %	77.5	((-160) to (-162.3) w/TFA
0.015% (930 µM)	64.7	6.96 %	N/A	(-119.4) to (-120.65)
0.008% (465 µM)	51.09	11.0 %	N/A	(-119.4) to (-120.65)
0.004% (232 µM)	26.04	11.2 %	N/A	(-119.4) to (-120.65)

DISCUSSION

Determining cytotoxicity of materials often utilizes dermal fibroblasts in media containing the test material and has been used to study silver ions. Silver nano particles have been shown in rat studies to accumulate in various organs from highest to lowest: lungs, spleen, liver, kidney, thymus, and heart [33]. Some observed effects include tissue irritation, chromosome breakage with potential genotoxicity and carcinogenic effects [34]. The use of silver nano particles could be further examined due to the concern that the public has previously expressed over the effect of long-term exposure to silver dental amalgam restorations [35]. Although the use of mercury in silver dental amalgam restorations was always stated as the chief concern, the anecdotal symptoms reported by some individuals may have been related to the release of nano silver particles [36, 37]. Silver, either in colloidal form, or ionic, or as nano particles has been associated with marked cytotoxicity [38]. Dermal fibroblasts exposed to silver have hindered proliferation and that has been associated with the delayed healing seen with anti-microbial products using silver particles [39, 40]. The very strong inhibitory effect of SDF on the dermal fibroblasts may have been the result of the combined cytotoxicity of both fluoride and silver. Unfortunately, we did not analyze the silver ion concentrations within the dermal cells, and most likely the precipitation that occurred in the media contained the silver ions. Plus, the concentration of the fluoride within the cells mostly matched the levels of SDF solution which correlated with the cytotoxicity observed. The international standard ISO 7405, which addresses the biocompatibility of dental materials and devices, states that biocompatibility tests should be performed on materials in an “as-used state” [41]. The amounts of SDF used by practitioners, especially non dental trained clinicians, may well exceed the amount expected with proper following of the manufacturer’s instructions (Six sites and air dry. For use in adults over the age of 21) [42]. The cytotoxicity studies used adults and not little children who may not cooperate with the SDF placement, allowing for soft tissue contact or even ingestion of the product. In addition, some practitioners place a Fluoride varnish on top of the SDF without knowledge of the potential toxicity of the combined treatment and with no evidence for increased effectiveness in decay prevention [43].

The Cytotoxicity of SDF on DF cells, HeLa cells and bacteria was clear in a previously completed study by the authors. This study was designed to try to quantify the actual cytotoxic endpoint of SDF. The results of this study further illustrate the scope of the problem in that the uncertainty persists. First, SDF when placed in a complex matrix like DF culture media, generates different fluoride containing components probably by associating with media proteins and other media components even after significant loss of SDF due to precipitation. Secondly, while NMR generates very concise values in soluble water matrix, the same is not true even in a simple saline or buffered matrix because of the loss of some material due to precipitation in these matrices even though the degree of precipitation is less. Finally, the use of internal fluorine containing standards alters the NMR signals (somewhat expected) so that one must plan future studies so that the internal standards are used as a template but not incorporated in the SDF containing samples to be quantified. This will still result in estimated values but is better when the other confounding variables are limited.

Considering that the SDF reagent is applied orally at a roughly 38% concentration (24.4-28.8% (w/v) Ag, 5.0-5.9% fluoride, at pH 10 and 8-10% ammonia) in aqueous solution, these results warrant more refined testing to identify true lower limit of toxicity end points of SDF. SDF should be utilized only by trained professionals and never contact soft tissue due to toxicity concerns. The TF should be considered from all sources; the public water, to dietary sources, from ingested toothpaste, professionally applied fluorides, both varnishes and SDF. The concern should not just be fluorosis, the risk we all understand, but rather neurological status, as in average IQ. The problem is that standardized weighted mean difference in IQ score between exposed and reference populations was -0.45 (95% confidence interval: -0.56, -0.35) using a random-effects model [44]. Thus, this study concluded that children in high-fluoride areas had significantly lower IQ scores than those who lived in low-fluoride areas. The potential for neurologic damage far out-weighs dental concerns [45]. Of interest is that, at concentrations higher than 0.7 parts per million, a systematic review found a low to moderate level-of-evidence that suggests adverse effects on learning and memory in animal models exposed to fluoride. The evidence is strongest (moderate level-of-evidence) in animals exposed as adults and weaker (low level-of-evidence) in animals exposed during development [46]. So what is the effect on humans? 21 of 23 recent epidemiological studies report an association between high fluoride exposure and reduced intelligence. This discrepancy between experimental and epidemiological evidence may be the result of deficiencies inherent in most of these epidemiological studies on a potential association between fluoride and intelligence, especially with respect to other confounding factors, e.g., breast feeding, socioeconomic status, maternal intelligence, residence, low birth weight, and exposure to other possible neurotoxic chemicals. In conclusion, based on the “totality of currently available scientific evidence, the present review does not support the presumption that fluoride should be assessed as a human developmental neurotoxicant at the current exposure levels in Europe” [6]. This research review article may not have considered the full effects of swallowed fluoride dental preventatives. This would present as a serious and obvious error. Indeed, the “most common sources of acute overexposures today are dental products–particularly dentifrices because of their relatively high fluoride concentrations, pleasant flavors, and their presence in non-secure locations in most homes. For example, ingestion of only 1.8 ounces of a standard fluoridated dentifrice (900-1,100 mg/kg) by a 10-kg child delivers enough fluoride to reach the ‘probably toxic dose’ (5 mg/kg body weight).” [47]

It has been reported that with a fluoride concentration of 44,800 parts per million, SDF is nearly twice the strength of commercially available 5% sodium fluoride varnishes used in primary care. Application of both agents on the same day is to be avoided as the fluoride dose would be additive, and its safety profile in children is still unknown [21]. However, some clinicians are apparently applying both at the same visit. Notwithstanding, Silver Diamine Fluoride (SDF) 38% has been receiving a great deal of attention by U.S. dental professionals since it was cleared for use by the Food and Drug Administration in August 2014 under the provisions of the Federal Food, Drug and Cosmetics Act. The Cleared Indications for Use are for the “Treatment of dentinal hypersensitivity. For use in adults over the age of 21.” Even though originally approved for only over the age of 21, this product is widely used on little children and even endorsed by major dental organizations as an off-label use. This is somewhat unexpected due to the reported cytotoxicity of the product [48]. One should assume a ‘cautionary approach” due to the vulnerability of young children. Even though, at “concentrations of 30% and 38%, SDF shows potential as an alternative treatment for caries arrest in the primary dentition and permanent first molars.” [49] Dental caries is, of course, a major concern, but so should be neurologic health and the lack of negative side-effects. There is an anti-microbial effect, “fluoride-treated tooth surfaces inhibit bacterial acid production at the bacteria/tooth interface. The SDF-coated RD had the strongest inhibitory effect.” [50] This may also imply an anti-mitochondrial effect, especially with males. Interestingly, a recent publication states “research confirms that SDF has long lasting antimicrobial effects at very low concentrations although it does raise concerns regarding cytotoxicity.” [51]

Dental prevention does not rely only on fluoride, but also on diet modification and the reduction in refined carbohydrates creates improved systemic health [52]. Processed sugar has been linked to diabetes, obesity, increased blood pressure, and even cancer [53, 54]. Obviously, dental professionals should discuss nutrition with all their patients, first and foremost. In addition, sugar overwhelms the protective effect of fluoride supplementation [55]. Hence leaving only the potential cytotoxicity and none of the dental preventive benefit for the patient. Other options are also available, such as, probiotic supplementation, which arguably is superior in preventive benefit to the patient [56, 57]. Of course, fluoride amounts could safely be reduced as long as the toothpaste, mouthwash or varnish had xylitol added to increase the effectiveness of the preventative [58, 59, 60, 61, 62].

In summary, SDF has an important role to play in the managing of dental caries and should be utilized by dental professionals as directed in the manufacturer’s instructions for use. The overall concern should be the TSF consumed by the young child. More of anything is not always better when treating dental patients, especially children. Dental professionals should also utilize SDF carefully preventing any ingestions or soft tissue contact and be aware of the TSF amounts the child patient is exposed to.

CONCLUSION

SDF has the potential for cytotoxicity and should be applied only by a trained dental professional according to the manufacturer’s indications. SDF application should be an important part of Preventive Dentistry but not the only dental preventive utilized. The practitioner should also consider the total systemic fluoride exposure for each patient, especially young children, and be cognizant of potential cytotoxicity.

Acknowledgement:

We are grateful to Dr. Yongbo Zhang PhD at Northwestern University’s IMSERC group for performing the NMR assessments and interpretations of the data included in this study.