3.4.1. Oatmeal Bar

Percent polymeric color values of oatmeal bar samples increased over storage at 4.4 ◦C (*p* = 0.0012) and 21 ◦C (*p* = 0.0173). For samples stored at 4.4 ◦C, % PC values increased from 7% at day 0 to 9.6% at eight weeks, and from 5.7% at day 0 to 9.6% at eight weeks in samples stored at 21 ◦C. Percent polymeric color values showed significant inverse correlations to levels of total anthocyanins at 4.4 ◦C (rxy = −0.65; *p* = 0.0081) and at 21 ◦C (rxy = −0.64; *p* = 0.0103) during storage.

#### 3.4.2. Graham Cracker Cookie

Percent polymeric color values of graham cracker cookie samples significantly increased over storage at 4.4 ◦C (*p* < 0.0001), but not at 21 ◦C (*p* = 0.2075). For samples stored at 4.4 ◦C, % PC values increased from 2.5% at day 0 to 6.3% at eight weeks, but when stored at 21 ◦C, % PC values remained at 6.2% on average during eight weeks of storage. Percent polymeric color values showed significant inverse correlations to levels of total anthocyanins at 4.4 ◦C (rxy = −0.56; *p* = 0.0298) and 21 ◦C (rxy = −0.58; *p* = 0.0237) during storage.

#### 3.4.3. Juice

Percent polymeric color values of juice samples significantly increased over storage at 4.4 ◦C (*p* = 0.0008) and 21 ◦C (*p* < 0.0001). For samples stored at 4.4 ◦C, % PC values increased from 13.6% at day 0 to 20.8% at eight weeks, and the values increased from 13.8% at day 0 to 23.5% at eight weeks when stored at 21 ◦C. Percent polymeric color values showed significant inverse correlations to levels of total anthocyanins at 4.4 ◦C (rxy = −0.58; *p* = 0.0239), and more strongly at 21 ◦C (rxy = −0.71; *p* = 0.0029) during storage.

#### 3.4.4. Gummy Product

Percent polymeric color values of gummy samples significantly increased over storage at 4.4 ◦C (*p* = 0.0229) and 21 ◦C (*p* < 0.0001). For samples stored at 4.4 ◦C, % PC values increased from 6.8% at day 0 to 12.7% at eight weeks, and in samples stored at 21 ◦C, % PC values increased from 6.9% at day 0 to 23.8% at eight weeks of storage. Percent polymeric color values showed a moderate inverse correlation to levels of total anthocyanins at 4.4 ◦C (rxy = −0.563; P = 0.036), but a much greater inverse correlation at 21 ◦C (rxy = −0.93; *p* < 0.0001) during storage.

Percent polymeric color values typically show an inverse correlation with total anthocyanins during storage of blueberry products [23,24], and inverse correlations with each individual anthocyanins in all the products and storage temperature (data not shown). Higher percent polymeric color values indicate that a higher percentage of anthocyanins are resistant to bleaching in the presence of potassium metabisulfite. Since the sulfonic acid adduct attaches at C4 on the middle heterocyclic ring, it is thought that anthocyanin–procyanidin polymers are formed via a direct condensation reaction, resulting in a C4–C8 anthocyanin–procyanidin linkage as the major polymers formed in blueberries during storage. Hence, it is possible that declines in anthocyanins during storage of the blueberry products are not true losses due to degradation, but the conversion of monomeric anthocyanins to anthocyanin–procyanidin polymers. Anthocyanins can be degraded via a hydration reaction, where the flavylium ion is converted to a hemiketal structure, which is rapidly converted to *cis*-chalcone, which slowly arranges to a *trans*-chalcone structure [40]. The *trans*-chalcone structure is highly unstable and rapidly degrades to hydroxybenzoic acid derivatives [40]. However, we do not consider that this reaction was responsible for anthocyanin losses in the blueberry products over storage since we did not observe an increase in phenolic acid derivatives in our HPLC chromatograms at 280 nm (data not shown).

### *3.5. Stability of Chlorogenic Acid during Storage*

The stability of chlorogenic acid in the four blueberry products stored at 4.4 ◦C and 21 ◦C is shown in Figure 5. Chlorogenic acid was stable in all products over storage regardless of storage temperature, except for the juice and oatmeal bar stored at 4.4 ◦C, where levels significantly decreased (*p* = 0.0249 and *p* = 0.0133, respectively). At 4.4 ◦C, the chlorogenic acid content decreased from 4.3 to 3.6 mg/g WBB powder in the juice and from 3.0 to 2.6 mg/g WBB powder in the oatmeal bar. At 21 ◦C, chlorogenic acid in the juice showed a slight increasing trend; however, this change was not statistically significant (*p* = 0.6496). Chlorogenic acid was also stable in the ice pop over eight weeks of storage at −20 ◦C (*p* = 0.8332), with an average value of 6.5 mg/g WBB powder over storage (Figure 1). Initial levels of chlorogenic acid were higher in all products stored at 21 ◦C compared with 4.4 ◦C storage, which may be due to the variation in processing the two sets of samples for the storage study, or possible the degradation of chlorogenic acid in the WBB powder used to prepare the products. The WBB powder used to prepare samples for the refrigerated storage study was stored at 15.5 ◦C for three months prior to preparing the samples. Blueberries contain polyphenol oxidase, which can readily oxidize chlorogenic acid [41]. Chlorogenic acid was previously found to be stable in blueberry juice, puree, and canned berries stored for six months at 25 ◦C [23], but blueberry jams lost 27% of chlorogenic acid over six months of storage at 25 ◦C [24].

**Figure 5.** Stability of chlorogenic acid in blueberry stored at 21 ◦C and 4.4 ◦C (*n* = 3/time point). Shaded area around lines represents 95% confidence intervals for predicted values.

#### *3.6. Stability of Flavonols in Blueberry Products during Storage*

The stability of total flavonols in the five blueberry products is shown in Figure 6. Total flavonol levels in the oatmeal bar stored at both temperatures and in the gummy product and graham cracker cookie stored at 4.4 ◦C were stable over eight weeks of storage as well as the juice samples stored at 4.4 ◦C from two to eight weeks (*p* > 0.05). Flavonols in the juice stored at 21 ◦C were also stable over time despite an upward trend, but the increase was not significant (*p* = 0.1935). Consistent with our findings, total flavonol concentrations were found to be relatively stable (<15% losses) in blueberry jam, juice, puree, and canned berries over six months of storage at 25 ◦C [24,42].

However, total flavonol content significantly decreased over storage in the gummy product (*p* = 0.0085) and graham cracker cookie (*p* = 0.0237) stored at 21 ◦C. Total flavonol levels declined by 45.7% and 28.5%, respectively, in these products over eight weeks of storage. In the gummy product, the most marked loss occurred from six to eight weeks of storage. Moisture loss during late storage presumably led to hardening of the gummies, resulting in incomplete extraction of the flavonols.

**Figure 6.** Stability of total flavonols in blueberry stored at 21 ◦C and 4.4 ◦C (*n* = 3/time point). Shaded area around lines represents 95% confidence intervals for predicted values.

(6" Individual flavonols in oatmeal bar and juice were all stable for both storage temperature as well as in graham cracker cookie stored at 4.4 ◦C (data not shown). In the graham cracker cookie stored at 21 ◦C, myricetin-3-galactoside was the most unstable flavonol showing 57.8% retention after eight weeks (*p* = 0.038). Quercetin-3-rutinoside (*p* = 0.0225), quercetin-3-galactoside (*p* = 0.0195), quercetin-3-glucoside (*p* = 0.037), quercetin-3-rhamnoside (*p* = 0.0058), quercetin-3-(6 ′′ -acetyl) galactoside (*p* = 0.0195), and syringetin-3-galactoside/glucoside (*p* = 0.0237) were all retained at levels from 61.1% to 70.1%. Four individual flavonols significantly decreased in the gummy product stored at 4.4 ◦C. The percent retention of quercetin-3-glucuronide (*p* = 0.0012) and quercetin-3-rhamnoside (*p* = 0.0016) were 33% and 38.5%, respectively. In the gummy product stored at 21 ◦C, only two flavonols (myricetin-3-glucoside and the syringetin derivative) showed no significant decrease over eight weeks of storage. All other flavonols in the gummy product stored at 21 ◦C were adversely affected by storage time with retentions ranging between 45.2 and 60.1% after eight weeks.

#### **4. Conclusions**

The stability of polyphenolics over eight weeks of storage in food products made with WBB powder varied according to product type. Polyphenolic compounds from the ice pop, oatmeal bar, and juice were shown to be stable over storage and are good candidates for further use in applications in which stored food items are to be used for delivery of significant amounts of polyphenolics (e.g., controlled feeding trials). The other food items may also be used for these applications, but the relative retention of bioactive polyphenolics as outlined herein should be taken into account when dose and delivery are designed. The gummy product showed relatively poor retention of anthocyanins and flavonols after eight weeks of storage, which may be due to extraction issues rather than true losses. In summary, incorporating WBB powder into food products in which key molecules remain intact during storage can improve the consumption of blueberry phytochemicals. The effect of the product matrix on bioavailability of retained polyphenolic compounds needs further investigation.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2304-8158/9/4/466/s1, Figure S1: Individual anthocyanin content of blueberry ice pop stored at −20 ◦C over eight weeks. Bars represent standard error of the mean (*n*=3/time point). Anthocyanidins: Cyd=cyanidin, Dpd=delphinidin, Mvd= malvidin, Pnd = peonidin, Ptd = petunidin. Sugar moiety: ara = arabinoside, gal = galactose, glc = glucose. Figure S2: Individual anthocyanin content of blueberry oatmeal bar stored at 4.4 and 21 ◦C over eight weeks. Bars represent standard error of the mean (*n*=3/time point). Anthocyanidins: Cyd=cyanidin, Dpd=delphinidin, Mvd= malvidin, Pnd = peonidin, Ptd = petunidin. Sugar moiety: ara = arabinoside, gal = galactose, glc = glucose. Figure S3: Individual anthocyanin content of blueberry graham cracker cookie stored at 4.4 and 21 ◦C over eight weeks. Bars represent standard error of the mean (*n* = 3/time point). Anthocyanidins: Cyd = cyanidin, Dpd = delphinidin, Mvd = malvidin, Pnd = peonidin, Ptd = petunidin. Sugar moiety: ara = arabinoside, gal = galactose, glc = glucose. Figure S4: Individual anthocyanin content of blueberry juice stored at 4.4 and 21 ◦C over eight weeks. Bars represent standard error of the mean (*n*=3/time point). Anthocyanidins: Cyd=cyanidin, Dpd=delphinidin, Mvd= malvidin, Pnd = peonidin, Ptd = petunidin. Sugar moiety: ara = arabinoside, gal = galactose, glc = glucose. Figure S5: Individual anthocyanin content of blueberry gummy stored at 4.4 and 21 ◦C over eight weeks. Bars represent standard error of the mean (*n*=3/time point). Anthocyanidins: Cyd=cyanidin, Dpd=delphinidin, Mvd= malvidin, Pnd = peonidin, Ptd = petunidin. Sugar moiety: ara = arabinoside, gal = galactose, glc = glucose.

**Author Contributions:** Conceptualization, L.H., S.H.A., J.-R.C., and E.C.D.; Methodology, L.H., C.B., and S.H.A.; Formal analysis, L.H., L.L., C.B., D.R., and A.M.; Investigation: L.H., C.B., and D.R.; Resources, L.H.; Writing—original draft, L.H., and L.L.; Writing—review and editing, L.H., L.L., S.H.A., J.-R.C., E.C.D., and A.M.; Supervision, L.H., and S.H.A.; Project administration, L.H., and S.H.A.; Funding acquisition, L.H., S.H.A., J.-R.C., and E.C.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by the Arkansas Children's Nutrition Center (USDA-Agricultural Research Service Project 6026-51000-012-06S), Little Rock, AR, USA. Eva C. Diaz is partially supported by the Arkansas Center for Advancing Pediatric Therapeutics funded by the National Institutes of Health (Award # 8UG1OD024945).

**Conflicts of Interest:** The authors declare no conflict of interest.
