Improved in vitro Hemagglutination Assays Utilizing P-Type and Type 1 Uropathogenic Escherichia coli to Evaluate Bacterial Anti-Adhesion Activity of Cranberry Products

Abstract

Cranberries have a long history of use in the prevention of urinary tract infections. Cranberry products vary in proanthocyanidin content, a compound implicated in preventing the adhesion of uropathogenic Escherichia coli (E. coli) to uroepithelial cells. Testing is routinely done by cranberry product formulators to evaluate in vitro bacterial anti-adhesion bioactivity, shelf-life, and potential efficacy of cranberry products for consumer use to maintain urinary tract health. Hemagglutination assays evaluate the anti-adhesion bioactivity of cranberry products by determining how effectively the products prevent agglutination of specific red blood cells with E. coli expressing P-type and Type 1 fimbriae. The current study sought to improve upon an established anti-adhesion assay method by expanding the number of E. coli strains used to broaden potential in vivo efficacy implications and presenting results using photomicrographic data to improve accuracy and build databases on products that are routinely tested. Different lots of cranberry powder ingredient and two formulated products were tested independently for anti-adhesion activity using the established method and the improved method. Positive harmonization of results on the same samples using rigorous controls was achieved and provides the substantiation needed for the cranberry industry to utilize the improved, rapid in vitro testing method to standardize cranberry products for sufficient anti-adhesion bioactivity and maintain consumer confidence.

Keywords:

• Bacterial anti-adhesion
• hemagglutination assay
• urinary tract
• cranberry
• proanthocyanidins

View correction statement:

Correction

Introduction

Cranberry products (CP) are extensively consumed worldwide for their beneficial effects on the prevention of UTI. A recent large-scale, comprehensive meta-analysis of 50 clinical trials published in The Cochrane Database of Systematic Reviews (Williams et al. 2023) found that consumption of certain CP can prevent and reduce the risk of recurrent UTIs in women, children, and other at-risk individuals. These benefits are substantially due to a group of condensed tannin oligomers and polymers of monomeric flavan-3-ols in cranberry called A-type proanthocyanidins (PACs), the predominant PACs (∼95%) in cranberries and their metabolites which have been shown in previous studies to prevent uropathogenic Escherichia coli (UPEC) bacteria that cause the majority of UTIs from adhering to bladder cells (Howell et al. 1998; Foo et al. 2000a, 2000b; Krueger et al. 2013; Feliciano et al. 2014; Mena et al. 2017; Iannuzzo et al. 2022). This mechanism for preventing UTIs is aimed at interrupting the first step in the urinary tract infection process (Beachey 1981), and since the bacteria are not killed, there is no selection pressure to develop resistance to the mechanism (Klinth et al. 2012).

Bacterial anti-adhesion assays have been developed based on this mechanism to allow manufacturers of CP, especially dietary supplements to evaluate the efficacy of the active compounds in their products that are targeted at maintaining urinary tract health. There are some limitations to these assays that have been used in the past to measure anti-adhesion activity (AAA), especially uroepithelial cell assays that count uropathogenic E. coli (UPEC) bacteria adhered to bladder cells, but because the cells have multiple adhesin receptors, these assays lack specificity for attributing adhesion prevention to either P-type or Type 1 E. coli adhesins (Neter 1956; Zafriri et al. 1989), the two most important virulence factors in UTI pathogenesis. P-type fimbrial adhesin is implicated in about 80% of pyelonephritis cases that cause kidney infections and up to 60% of cystitis cases, and Type 1 fimbrial adhesin in over 70% of cystitis cases (Johnson 1991; Foxman 2002; Wullt 2003). Adhesion of both P-type UPEC and Type 1 UPEC are inhibited by cranberry (Di Martino et al. 2006; Rafsanjany et al. 2015; Liu et al. 2019), especially A-type PACs (Foo et al. 2000a; Lavigne et al. 2008), so it is important that assays on CP have specificity for detecting AAA against each of these virulence factors. Validated hemagglutination assays using microscopy for evaluation are able to distinguish differences in bacterial adhesin binding specificity to cell surfaces (Mrazkova et al. 2019) and have been successfully utilized to detect inhibition of red blood cell (RBC) agglutination by extracts of edible plants and isolated compounds (Eltigani et al. 2019). Bacterial-mediated hemagglutination has been extensively used as a surrogate marker for bladder cell adhesion, due to its specificity for determining activity against discreet adhesins (Evans et al. 1977; Hagberg et al. 1981; Väisänen et al. 1981). RBC agglutination occurs when E. coli fimbriae bind to the surface of blood cells. P-fimbriated E. coli (PapG) induce human RBCs (A + or O + Rh factor) to agglutinate, and Type 1-fimbriated E. coli (FimH) induce agglutination of guinea pig RBCs (Hagberg et al. 1981). P-type E. coli recognizes and binds to a specific galactosyl-galactose structure on bladder cells and human RBCs which is part of the P blood group antigens, and Type 1 E. coli recognizes and binds to mannose-like structures (Klinth et al. 2012).

Important considerations for improving the accuracy of measuring AAA in hemagglutination assays involve the composition of the CP and the type, levels and stability of the PACs found in the products. It is known that there is substantial variability in composition and efficacy among CP used by consumers for UTI prevention (Krenn et al. 2007; Sánchez-Patán et al. 2012; Chughtai et al. 2016; Mannino et al. 2020; Williams et al. 2023), so proper standardization and analytic testing of raw ingredients and finished CP is critical to ensure accuracy in the AAA testing. Harvest and storage of cranberries and subsequent harsh processing, especially at high heat or oxidizing conditions can negatively impact the PAC molecular structures and bacterial anti-adhesion bioactivity of cranberry in CP products (Pappas and Schaich 2009). Further, not all cranberry fruit components used to formulate CP possess anti-adherence properties. Only the water-soluble PACs in CP derived from cranberry juice and juice extracts exhibit substantial anti-adhesion bioactivity compared to the insoluble PACs found in the skins and pulp of cranberry which are bound up with cellulose in some CP, reducing their efficacy (Howell et al. 2022). Therefore, it is important to accurately measure levels of soluble PAC using the aldehyde condensation of 4-(dimethylamino)cinnamaldehyde (DMAC) assay (Krueger et al. 2013; Sintara et al. 2018; Birmingham et al. 2021), and then relate those levels to the anti-adhesion bioactivity to ensure that the PACs have maintained their potency and accurately reflect the potential benefits of the CP for maintaining urinary tract health.

Dr. Amy Howell’s laboratory at Rutgers University adapted and has utilized the microscopy-based hemagglutination assay continuously for over 20 years to determine AAA of hundreds of samples of whole cranberry, its components, fractions, isolated PACs and finished CP (especially dietary supplements) (Foo et al. 2000a, 2000b; Howell et al. 2005; Howell et al. 2010; Howell et al. 2015; Howell et al. 2022). This anti-adhesin assay is widely utilized by the cranberry industry both in the United States and internationally. It provides suppliers of raw ingredients and manufacturers of finished consumer products assessment of in vitro AAA of their cranberry materials for bioactivity, standardization, and shelf-life determination. In many cases, each lot is tested and results appear on product spec sheets. The assay measures the capacity of the cranberry components to inhibit agglutination (a surrogate marker for adhesion) of E. coli (P-type or Type 1) with specific blood cell types. The cranberry sample is serially diluted (2-fold), resulting in a range of concentrations of the sample from high to low. The bacteria are then pre-incubated with each dilution of the cranberry sample prior to incubation with specific RBCs. Each dilution is then visually examined microscopically for the degree of hemagglutination. The dilution concentration of the cranberry sample at which 50% hemagglutination is present is considered the minimum inhibitory concentration (MIC) anti-adhesion bioactivity value and is reported on the supplier’s spec sheets. This is similar to a measure of half maximal inhibitory concentration (IC50) to determine the potency of a substance in inhibiting a specific biological or biochemical function. Therefore, a low MIC value represents a high level of AAA.

The Rutgers method effectively quantifies in vitro AAA of water-soluble PACs in CP (Gupta et al. 2007; Chughtai et al. 2016) and was expanded to determine the qualitative ex vivo anti-adhesion (+ or −) of urine collected following discreet intakes of CP by human participants (Howell et al. 2005; Howell et al. 2015; Howell et al. 2022). In these studies in which the determination of in vitro AAA of CP was followed up with controlled CP intake and human urine collection trials over time, it was possible to detect ex vivo AAA trends in urine correlating with intake doses. Urine samples collected over the various time points were tested for AAA following incubation with UPEC at 100,000 CFU/mL (the concentration used to diagnose clinical UTI), allowing the urine assay to more closely mimic the in vivo conditions in the bladder when UPEC are present at infection-causing concentrations. This has enabled in vitro AAA results on CP containing water-soluble PAC from juice extracts to be assigned into categories based on the strength of the AAA (low, medium or high) and could potentially help estimate in vivo efficacy when the CP is ultimately consumed.

Rutgers University ended testing of the anti-adhesion bioactivity of CP in early 2023 and transferred the hemagglutination assay methods to Complete Phytochemical Solutions (CPS), LLC in Cambridge, WI. The main objective of the current study was to run the assays on identical CP samples at both Rutgers and CPS to ensure that results could be replicated. Multiple samples of a highly standardized juice-based cranberry powder ingredient and two finished products formulated from the powder were chosen for this harmonization study because Rutgers had accumulated 17 years of AAA data on them with extremely consistent results. Even though the fundamental assay methods are the same, a number of improvements to the Rutgers’ method (previously described) (Foo et al. 2000a; Howell et al. 2005; Howell et al. 2022) were undertaken by CPS. These included expanding the UPEC strains used in the hemagglutination assay to include both P-type (which was emphasized in Rutgers testing for many years) and Type 1 (which has become the focus of many studies), as well as presentation of photomicrographic data of MIC endpoint values to allow for greater accuracy in determining results. Here we report the results of the harmonization of methods between labs and improvements incorporated by CPS. Positive harmonization will provide the substantiation needed for the cranberry industry to utilize the improved, rapid in vitro testing method to standardize CP for sufficient anti-adhesion bioactivity. This will help consumers have confidence in the efficacy of CP they consume for maintaining urinary tract health.

Materials and methods

Description of standards and CP samples

Gikacran (Pharmatoka S.A.S., France) is a refined dried cranberry juice extract, standardized to 17.5% ±5% (w/w) PACs reported as Procyanidin A2 dimer equivalents using the DMAC quantitation assay. Due to rigorous PAC standardization and highly consistent hemagglutination results for over a decade, Rutgers routinely used Gikacran as an internal standard in all anti-adhesion assays to evaluate AAA of CP. A research standard of similar characteristics/composition can be purchased from USP (USP Cranberry fruit juice dry Extract RS: https://store.usp.org/product/115021). Successful harmonization of the Rutgers results with those from CPS would validate the use of Gikacran as a standard in CPS anti-adhesion testing moving forward. To confirm this, five production lots of Gikacran (43-JKC-21T0061700, 43-JKC-21T0061800, 43-JKC-21T0061900, 43-JKC-21T0062200, and 43-JKC-21T0062300) were evaluated for AAA in the current study. Pure cranberry PAC was isolated from cranberry fruit as previously described (Feliciano et al. 2014), and used as a positive control to confirm that AAA of the PACs could be detected using the hemagglutination assays. The MIC of pure PAC is considered to be the highest in vitro AAA value, to allow relative comparisons to MICs of the tested CP samples and internal cranberry powder standard (Gikacran).

Ellura and Urell are two CP brands of herbal medicines and medical-grade food supplements developed by Pharmatoka S.A.S (France) that are formulated with Gikacran. The recommended serving for Ellura and Urell is 1 capsule (265 mg/capsule) containing 206 mg (average 196–217 mg) Gikacran standardized to deliver 36 mg PAC (Procyanidin A2 equivalents). Two production lots of Ellura herbal medicine (N0205803 and M0161602) and 3 production lots of Urell food supplement and herbal medicine (P016701, M0754501, and M0127402) were evaluated in this study to harmonize the CPS and Rutgers methods and improve accuracy and presentation of results.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS)

Comparison of the PACs structural features in Gikacran and Ellura/Urell PAC was performed to ensure the PAC were similar to authentic cranberry with typical A-type linkages. Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) analysis was performed following AOAC First Action Method: 2019.05 (Esquivel-Alvarado et al. 2021b). Individual samples were extracted and prepared for analysis with a Bruker autoflex^® maX MALDI-TOF MS equipped with a SmartBeam II laser (Billerica, MA, USA). MALDI-TOF MS conditions: ion source 1 (19.0 kV), ion source 2 (16.70 kV), lens voltage (8.4 kV), reflector 1 (21.0 kV), reflector 2 (9.55 kV), pulsed ion extraction (120 ns), detector gain (4.0; 1910 V), preamplifier filter bandwidth (ultra), and digitizer sampling frequency (2000 Hz). Samples and matrix (DHB; 100 mg/mL MeOH) were mixed at a 50:50 ratio and spotted (0.5 μL) on the MALDI-TOF MS stainless steel target. Spectra were calibrated using a cranberry PAC reference standard (Esquivel-Alvarado et al. 2021b). Spectra are the sum of different locations in each well, accumulating a total of 2000 shots with deflection set at 580 Da. Data was acquired on three technical replicates (individual wells on the MALDI target) for each PAC sample. FlexControl and Flex Analysis software (Bruker Daltonik GmbH, Bremen, Germany) were used for data acquisition and data processing, respectively.

High performance liquid chromatography analysis of purified cranberry fruit proanthocyanidins

HPLC analysis was performed on the pure cranberry PAC standard to confirm that the preparation was enriched in PAC and free from other classes of phenolic compounds (anthocyanins, hydroxycinnamic acids and flavonols). A Waters HPLC (e2695 Separations Module) equipped with a PDA detector (Waters, 2998) was used with an Agilent Zorbax SB-C18 HPLC column (4.5 mm × 150 mm, 5 µm) for analysis. The column was set to a temperature of 40 °C, and a flow rate of 1.0 mL/min was used. The gradient profile was as follows: A 0.1% TFA in H₂O (HPLC Grade, Sigma Aldrich) solution was used as mobile phase A and 0.1% TFA in methanol (HPLC Grade, J. T. Baker) solution was used for mobile phase B. 0–10 min, 90–70% A; 10–25 min, 70% A; 25–45 min, 70–45% A; 45–50 min, 45–10% A; 50–55 min, 10% A. A 10-minute equilibration step of 10 min at initial conditions was set before the sample set was initiated.

Sample and standard preparation

Studies have demonstrated that previously tested CP with a MIC of 60 mg/mL in in vitro anti-adhesion assays are at the minimum threshold for detection of ex vivo AAA in urine following consumption (Howell et al. 2022). Thus, CP that agglutinate in the in vitro assays at 60 mg/mL are considered negative with no AAA. This is why the dilution series on CP are initiated at a concentration of 60 mg/mL. Since cranberry PACs are the most active anti-adhesion component in CP with very low MIC thresholds for AAA (Howell et al. 2022), the assay dilution series of pure PAC begins at a lower concentration of 7.5 mg/mL. The MIC of pure PACs represents the highest AAA for relative interpretation of the results.

Ellura and Urell capsules were opened and the powdered contents used for analysis. Gikacran standard, Ellura and Urell samples were prepared by weighing 60 mg into separate microfuge tubes. Phosphate buffered saline (PBS; 1 mL) was added to each tube resulting in an initial concentration of 60 mg/mL. Tubes were vortexed to mix the contents and centrifuged 16,200 × g, 1 min) to pellet insoluble materials. Supernatants were collected and serially diluted 1:1 (v/v) with PBS to produce 12 concentrations (60 mg/mL–0.06 mg/mL) which were evaluated in the studies. A two-fold dilution series (7.5 mg/mL–0.015 mg/mL) was prepared from the pure cranberry PAC to serve as a positive anti-adhesion control for the assays and determine the highest AAA threshold.

Bacterial cultures

Escherichia coli expressing P fimbriae

A cryocare bead (Key Scientific) was loaded with a P-type UPEC strain originally isolated from a patient with pyelonephritis and designated as CPS001. The strain was removed from −80 °C storage and the bead rubbed on King’s Agar B plate (Crystalgen) to deposit the bacteria on the surface of the agar according to the recommended protocol (Key Scientific) and then incubated statically at 37 °C for 12 h. Culturing on King’s Agar B enhances the expression of P fimbriae. Following incubation, bacterial colonies were picked from the agar with a loop, suspended in PBS, and diluted to an optical density (OD) of 0.5 absorbance units (600 nm) which is approximately a 5 × 10⁸ bacterial concentration. This bacterial preparation was used in the comparative hemagglutination assays to detect AAA.

Escherichia coli expressing type 1 fimbriae

A cryocare bead (Key Scientific) loaded with a UPEC strain CFT073 (ATCC 700928) capable of expressing Type 1 fimbriae was removed from −80 °C storage and transferred to 15 mL tryptose broth (100 mm tissue culture plate). The plate was swirled to mix the bacteria with the broth and then incubated statically at 37 °C for 12 h. Culturing E. coli CFT073 in the 100 mm plate provides a large surface area for oxygen exchange with the tryptose broth, which enhances the expression of Type 1 fimbriae. Following incubation, bacteria were transferred to PBS and diluted to an OD of 0.5 absorbance units (600 nm) to achieve a 5 × 10⁸ bacterial concentration for use in the comparative anti-adhesion assays.

Hemagglutination assays to detect AAA

Five production lots of Gikacran, 2 lots of Ellura, and 3 lots of Urell were prepared for evaluation in the assays as described above. Each dilution was analyzed in duplicate within plates for both E. coli fimbrial types (P-type and Type 1). The hemagglutination assays were repeated on three different days. Sample dilutions and bacterial cultures were prepared each day of the experiment. This experimental design enabled us to show consistency in the determination of MIC within and across samples on a single plate and within and across samples on multiple days (n = 3). CPS results were compared to data generated at Rutgers University on the same samples to determine if harmonization was achieved.

Anti-adhesion assay (P-type E. coli)

Sample dilutions of individual lots of Gikacran standard, Ellura, Urell and pure PAC positive control were individually pipetted (90 μL) in duplicate into individual wells on 96-well V-bottomed plates. P-type E. coli bacterial suspension (5 × 10⁸ bacterial/mL PBS) was pipetted (30 μL) into each well. Plates were incubated at room temperature for 15 min on a plate shaker (1000 RPM) to allow samples to interact with bacteria. Following primary incubation, 3% whole human blood (WHB) (O, Rh+) (Innovative Research, Inc., Novi, MI) in PBS (v/v) was pipetted (30 μL) into each well. Red blood cells in WHB (O⁺ or A⁺) agglutinate P-type E. coli but not Type 1. Controls including WHB alone, WHB with E. coli, WHB with the sample, pure PAC control at 7.5 mg/mL (highest concentration), and E. coli with the sample, standard or pure PAC control were also prepared in individual wells on the plates. Plates were placed on a plate shaker (1000 RPM) and incubated for an additional 15 min at room temperature, followed by centrifugation to concentrate materials at the bottom of the well for microscopic evaluation (1000 × g, 1 min).

Anti-adhesion assay (Type 1 E. coli)

Sample dilutions of individual lots of Gikacran standard, Ellura, Urell, and pure PAC control were individually pipetted (90 μL) in duplicate into individual wells on 96-well V-bottomed plates. Type 1 E. coli bacterial suspension (5 × 10⁸ bacterial/mL PBS) was pipetted (30 μL) into each well. Plates were incubated at room temperature for 15 min on a plate shaker (1000 RPM) to allow samples to interact with bacteria. Following primary incubation, 3% guinea pig red blood cells (GPRBC) (Innovative Research, Inc., Novi, MI) in PBS (v/v) was pipetted (30 μL) into each well. GPRBC agglutinate Type 1 E. coli, but not P-type. Controls including GPRBC alone, GPRBC with E. coli, GPRBC with the sample, standard or pure PAC control at 7.5 mg/mL (highest concentration), and E. coli with the sample, standard or pure PAC control were also prepared in individual wells on the plates. Plates were placed on a plate shaker (1000 RPM) and incubated for an additional 15 min at room temperature, followed by centrifugation to concentrate materials at the bottom of the well for microscopic evaluation (1000 × g, 1 min).

Microscopy imaging

Following centrifugation, the 96-well plates were viewed under a microscope (Echo Revolve) with a 4× objective. Digital photomicrographic images of each dilution well were captured. Independently, scientists at both CPS and Rutgers visually determined the MICs of the samples, standards, and PAC control. MIC assessments were then compared between both laboratories to evaluate the harmonization of results.

Results

MALDI-TOF MS

Positive reflectron mode MALDI-TOF MS analysis showed that Gikacran and Ellura/Urell are enriched in a PAC oligomeric series from the trimer to the heptamer. Deconvolution of the MALDI-TOF spectra shows both Gikacran and Ellura/Urell contained PACs that have predominantly one or more A-type interflavan bonds at each degree of polymerization (Table 1). The relative A-type to B-type interflavan bond ratios support the authenticity of cranberry PAC and are consistent with previously published results (Esquivel-Alvarado et al., 2021a, 2021b; Feliciano et al., 2012). There is small variation in the estimation of A-type to B-type interflavan bonds at each PAC degree of polymerization, demonstrating consistency in manufacturing of Gikacran and formulation of Ellura/Urell.

HPLC results

HPLC analysis of the purified cranberry PAC standard showed that the preparation was enriched in PAC, evidenced by the detection of a poorly resolved ‘hump’ at 280 nm that is associated with the oligomeric distribution of PAC (Figure 1). The purified cranberry PAC were devoid of other classes of phenolic compounds; anthocyanins (520 nm), hydroxycinnamic acids (320 nm), and flavonols (370 nm).

Figure 1. HPLC chromatograms of purified cranberry PAC. The poorly resolved ‘hump’ at 280 nm wavelength shows an oligomeric proanthocyanidin distribution. The preparation is devoid of other polyphenols, as evidenced by the absence of individual peaks resolving from baseline; hydroxycinnamic acids (320 nm), flavonols (370 nm) and anthocyanins (520 nm).

Hemagglutination assays to detect AAA

Anti-adhesion assay (P-type E. coli)

The assay results of the positive and negative controls were as expected (Figure 2). Following centrifugation, the WHB negative control formed a tightly packed pellet centered at the bottom of the well with no agglutination present. The WHB + P type E. coli (WHB + EC) positive control successfully agglutinated, forming a pellet with amorphous boarders centered at the bottom of the well. In comparison to the WHB control alone, the amorphous nature of the WHB + P type E. coli pellet is a result of initial agglutination of blood cells followed by accelerated sedimentation by means of centrifugation. The WHB + PAC negative control at a concentration of 7.5 mg/mL did not agglutinate, exhibiting a diffuse deposition of the blood cells across the bottom and side walls of the well. The pure PAC at 7.5 mg/mL + P type E. coli (EC + PAC) negative control showed diffuse deposition across the bottom and side walls of the well, indicating no agglutination.

Figure 2. Representative images of P-type and Type 1 E. coli anti-adhesion activity (AAA) results expressed as minimum inhibitory concentration (MIC) of Gikacran standard, Ellura and pure cranberry PAC positive control using hemagglutination-based assays. The dilution series from 60 mg/mL–0.015 mg/mL is presented along the top row and the MIC results for each test sample are displayed in each column. Yellow star indicates resulting MIC dilution required to suppress hemagglutination of bacteria by 50% in each test run. Scale of relative AAA represents potential in vivo efficacy of samples. Representative P-type negative control images (left to right) whole human blood (WHB), WHB + CPS001 P-fimbriated E. coli, WHB + 7.5 mg/mL pure cranberry PAC control, and CPS001 P-fimbriated E. coli + 7.5 mg/mL pure cranberry PAC control. Representative Type 1 negative control images (left to right) guinea pig red blood cells (GPRBC), GPRBC + CFT073 Type 1-fimbriated E. coli, GPRBC + 7.5 mg/mL pure cranberry PAC control, and CFT073 Type 1-fimbriated E. coli + 7.5 mg/mL pure cranberry PAC control.

Table 1. Deconvolution of positive reflectron mode MALDI-TOF MS showing Gikacran (n = 5) and Ellura/Urell (n = 5) samples contain PAC oligomeric distribution from trimers through heptamers, with predominantly one or more A-type bonds at each degree of polymerization.

	Gikacran (n = 5)		Ellura/urell (n = 5)
	Average	STD	Average	STD
Trimers
2A-type:0B-type	7.3	2.3	5.1	1.4
1A-type:1B-type	91.6	2.0	93.7	1.2
0A-type:2B-type	1.1	0.4	1.2	0.7
Tetramers
3A-type:0B-type	1.5	0.2	1.2	0.3
2A-type:1B-type	9.2	1.1	8.1	1.7
1A-type:2B-type	87.5	1.4	89.4	2.5
0A-type:3B-type	1.9	1.2	1.3	0.6
Pentamers
3A-type:0B-type	1.8	0.1	1.6	0.4
2A-type:1B-type	17.7	1.4	17.8	3.8
1A-type:2B-type	75.3	0.9	76.2	3.5
0A-type:3B-type	5.2	2.0	4.3	1.1
Hexamers
3A-type:0B-type	3.2	0.5	3.5	1.0
2A-type:1B-type	45.6	3.0	47.3	6.1
1A-type:2B-type	45.2	2.5	43.1	5.6
0A-type:3B-type	6.0	1.1	6.0	1.6
Heptamers
3A-type:0B-type	10.7	0.7	11.6	2.4
2A-type:1B-type	53.1	5.5	51.2	5.6
1A-type:2B-type	29.0	5.3	30.8	5.1
0A-type:3B-type	7.2	2.1	6.4	3.4

The dilution series of pure PAC positive control exhibited P-type hemagglutination at a MIC of 0.03 mg/mL in a dose-dependent manner, indicating that the PACs were highly effective at inhibiting bacterial adhesion. The 0.03 mg/mL dilution was designated as representing the highest AAA for assessment of the relative AAA of the other samples using the scale in Figure 2. The MIC occurred at the point at which the pelleted blood first began to form a loose, centralized amorphous pellet (about 50% agglutinated) (see yellow star in Figure 2) and resembled the MIC results of the WHB + P type E. coli positive control. The 0.015 mg/mL dilution had a very tight pellet and was considered fully agglutinated, which was one dilution more than the target MIC of 50%.

All Gikacran standard, Ellura, and Urell samples inhibited hemagglutination of P type E. coli with WHB in a dose-dependent manner. The MICs of all 5 Gikacran samples were identical at 0.12 mg/mL over all reps and days tested in both Rutgers and CPS analyses. MICs of 2 Ellura and 3 Urell production lots in all assay runs were consistently 0.23 mg/mL when the assays were performed by Rutgers and CPS. Since there was no variation in the results between days or reps, Table 2 presents the average MIC values (0.12 mg/mL for Gikacran) (0.23 mg/mL for Ellura and Urell) of the tested lots.

Table 2. Comparison of minimum inhibitory concentrations (MIC) at which Gikacran cranberry standard and sample cranberry products prevent E. coli-associated hemagglutination of blood cells using the CPS and Rutgers anti-adhesion assay methods for P-type and Type 1 E. coli.

Sample lots	MIC (mg/mL)	MIC (mg/mL)
	P-type E. coli*	Type 1 E. coli**
	Method: CPS Rutgers	Method: CPS Rutgers
Gikacran 43-JKC-21T0061700	0.12	0.47
Gikacran 43-JKC-21T0061800	0.12	0.47
Gikacran 43-JKC-21T0061900	0.12	0.47
Gikacran 43-JKC-21T0062200	0.12	0.47
Gikacran 43-JKC-21T0062300	0.12	0.47
Ellura N0205803	0.23	0.47
Ellura M0161602	0.23	0.47
Urell P01670	0.23	0.47
Urell M0754501	0.23	0.47
Urell M0127402	0.23	0.47

MIC values were identical among all sample lot replicates across testing days, therefore one value (the average) is reported for each sample tested.

* Agglutination of red blood cells from whole human blood (O⁺).

** Agglutination of Guinea pig red blood cells.

To illustrate the procedure for visually determining MIC values (yellow stars in Figure 2) based on the hemagglutination response in the dilution series for controls, standards and CP tested, representative photomicrographic AAA results are presented from one of the Gikacran lots (43-JKC-21T0061700) and one Urell lot (N0205803). The lower MIC of Gikacran (by one dilution) compared to Ellura and Urell indicates that less Gikacran is required to produce hemagglutination, making it more active at preventing bacterial adhesion. This is expected as both Ellura and Urell, the consumer products formulated from Gikacran, contain excipients resulting in a lower concentration of PACs per unit weight than in the Gikacran ingredient alone.

Anti-adhesion assay (Type 1 E. coli)

The assay results of the positive and negative controls were as expected (Figure 2). Following centrifugation, the GPRBC negative control formed a tightly packed pellet centered at the bottom of the well with no agglutination present. The GPRBC + Type 1 E. coli positive control (GPRBC + EC) successfully agglutinated, forming a pellet with amorphous boarders centered at the bottom of the well. In comparison to the WHB + P type E. coli (WHB + EC) control, the amorphous nature of the GPRBC + Type 1 E. coli pellet was slightly more diffuse, showing greater distribution along the walls of the well. The GPRBC + PAC negative control at a concentration of 7.5 mg/mL did not agglutinate showing a diffuse deposition of the blood cells across the bottom and side walls of the well. The pure PAC + Type 1 E. coli (EC + PAC) negative control showed diffuse deposition across the bottom and side walls of the well, indicating no agglutination.

The dilution series of pure PAC positive control exhibited Type 1 hemagglutination at a MIC of 0.12 mg/mL in a dose-dependent manner, indicating that the PACs were highly effective at inhibiting bacterial adhesion. The 0.12 mg/mL dilution was designated as representing the highest AAA for assessment of the relative AAA of the other samples using the scale in Figure 2. The MIC occurred at the point at which the pelleted blood first began to form a loose, centralized amorphous pellet (about 50% agglutinated) (see yellow star in Figure 2) and resembled the MIC results of the GPRBC + Type 1 E. coli positive control. The 0.06 mg/mL dilution had a very tight pellet and was considered fully agglutinated, which was one dilution more than the target MIC of 50%.

All Gikacran standard, Ellura, and Urell samples inhibited hemagglutination of Type 1 E. coli with GPRBC in a dose-dependent manner. The MICs of all 5 Gikacran, 2 Ellura, and 3 Urell production lots over all testing days and reps were consistently 0.47 mg/mL in both Rutgers and CPS analyses, considered to have ‘medium’ AAA on the anti-adhesion scale in Figure 2. Therefore, the average MIC value over all days tested was also 0.47 mg/mL and is the value presented in Table 2 for the different production lots.

Since MIC values were identical, photomicrographic AAA results from one of the Gikacran lots (43-JKC-21T0061700) and one Urell lot (N0205803) were selected to visually represent the determination of the MIC value (yellow star) based on the hemagglutination response in the dilution series of each sample (Figure 2).

Discussion

Results of the current study support the positive harmonization of the Rutgers and CPS hemagglutination assays for determining AAA of CP. These results demonstrate consistency in the 96-well hemagglutination assay within a single day as evaluated by comparing technical replicates (n = 2) of each Gikacran standard, Ellura, and Urell production lot within the plate. There was also consistency in MIC values comparing each Gikacran production lot across multiple days (n = 3). Similarity across days shows that the variability in culturing of bacteria (new bacterial culture prepared daily) was not influencing the MIC. Other possible sources of variation including inaccurate weighing of CP samples for testing, inconsistencies in sample dissolution or reagent pipetting, if present, did not influence the data outcomes. In addition, the positive and negative controls, which are very important for reducing experimental error due to issues with the bioassay components, all behaved as expected, improving confidence in the accuracy of MIC values determined for the cranberry samples. The positive controls using RBCs and UPEC are vital for the detection of fimbrial expression in the different UPEC strains. P fimbrial expression was confirmed in CPS001 E. coli because it successfully agglutinated the WHB control, and Type 1 fimbrial expression was confirmed in CFT073 E. coli because it agglutinated the GPRBC control. The high AAA of the PACs was demonstrated by the prevention of adhesion when the pure cranberry PAC positive control was incubated with the different UPEC strains at very low concentrations. The inclusion of the Gikacran standard was important because it contains not only high levels of active PAC (17.5%) but the other phenolic compounds in cranberry juice (anthocyanins, flavonols, and hydroxycinnamic acids) which did not interfere with the assay results. It is more representative of other CP (juices, dried fruit, sauces, supplements) that are tested using the assay. Since there are years of consistent historical data on the AAA of Gikacran, it can confidently be used as a standard in anti-adhesion testing in the future using the CPS methods.

Following Type 1 E. coli anti-adhesion testing, MIC values for the Gikacran ingredient and the branded products Ellura and Urell (both with added excipients) were identical which suggests that the in vitro assay may not be quite as sensitive in detecting AAA differences among CP samples against Type 1 UPEC; however unpublished data from Rutgers indicates that when Ellura is consumed the ex vivo urinary AAA is high. There may be other beneficial mechanisms of action of Ellura against Type 1 E. coli that remain to be elucidated. Thus, there is value in utilizing the Type 1 assay when testing CP because it is useful for detecting product adulteration and assessing the AAA of individual production lots and shelf-life.

Determination of the MICs of each sample was facilitated by multiple analysts viewing the photomicrographs taken at each dilution to reduce the subjectivity in selecting the 50% hemagglutination well. In each case on the current samples, the dilution with 100% hemagglutination was easy to identify because the bacterial-RBC complexes were pelleted into a tight ball at the bottom of the well. The previous dilution was designated as the MIC. In the future when the assays are used to assess AAA on other types of CP samples, there may be very rare occasions when the RBC-bacterial complexes in one well are 25–50% agglutinated, and the subsequent wells are 50–75% agglutinated. In these cases, a MIC listing the range between the two values will be reported. The photographic evidence will be archived for all future samples tested, allowing reexamination of the results, if necessary. In addition, a results database will be created on each CP and used for comparing results on previous lots to allow for tracking of shelf-life and detection of any manufacturing or product quality issues that may be present in the new product samples.

The expansion of the CPS methods to include analysis of AAA against both Type 1 and P-type UPEC expands the usefulness of the results for predicting possible in vivo efficacy of CP to a broader range of infections, including bladder (cystitis) and kidney infections (pyelonephritis). The AAA scale in Figure 2, based on the correlation of previous in vitro AAA results with ex vivo urinary post-consumption AAA data (Howell et al. 2005; Howell et al. 2015; Howell et al. 2022) may be useful for estimating potential in vivo efficacy of products. Using the scale which ranges from a low of 60 mg/mL to the MIC of the pure PAC standard (0.03 mg/mL for P-type assay and 0.12 mg/mL for Type 1 assay), Ellura and Urell are assigned ‘high’ AAA against P-type and ‘medium’ for Type 1 E. coli. They are both formulated with Gikacran, a cranberry juice-based extract containing soluble PACs high in vitro and urinary ex vivo AAA, compared to cranberry pomace-based products in which the PACs are bound to cellulose and have reduced in vivo or ex vivo AAA (Howell et al. 2022). These pomace and skin-based CP are designated as having ‘low’ AAA on the scale, with MIC values typically between 60 and 15 mg/mL. Whole berry powders and other products that are a mix of juice extract and pomace historically (using the Rutgers method) have MIC values of 7.5–0.47 mg/mL and are considered ‘medium’ on the AAA scale.

Given the broad array of CP on the market formulated with different components that may or may not contribute to AAA (including adulterated products), it is important to have access to testing to determine AAA, an important aspect of product standardization. Quantitation of PAC levels in CP is also important for standardization, but PAC levels do not necessarily correlate directly with AAA. Since harsh processing conditions can impact PAC structures and certain oil and cellulose excipients in CP can bind up PACs, subsequent AAA of the finished product can be negatively affected. So, even though there may be high PAC content, the product may not have high AAA, making it difficult for consumers to know which products will be efficacious. This issue confirms the importance of running anti-adhesion testing on all CP. The CPS method can also detect AAA of products with other components, even if they are not pure cranberry.

The improved CPS method harmonized with the Rutgers method in this study successfully demonstrated consistency in manufacturing of the cranberry samples tested and established a specification for anti-adhesion bioactivity. This specification for the ingredient and blended finished products compliments quantification and authentication of cranberry PACs as part of quality management under guidelines for good manufacturing practice (cGMP).

Disclosure statement

Scott Bosley and Andrew Birmingham are employed by Complete Phytochemical Solutions, LLC, and Christian G. Kruger, Amy B. Howell and Jess D. Reed have ownership interest.

Correction Statement

This article was originally published with errors, which have now been corrected in the online version. Please see Correction: https://doi.org/10.1080/19390211.2023.2286740

Additional information

Funding

This study was supported by an intramural fund at Rutgers University and a research grant from Pharmatoka SAS (France).

Notes on contributors

Scott Bosley

Scott S. Bosley is a Scientist for Complete Phytochemical Solutions, LLC. Graduated from the University of Wisconsin – Madison in 2020 with a Master of Science in Molecular and Cellular Pharmacology. He is focused on developing assays to test the biological efficacy of dietary supplements.

Christian G. Krueger

Christian G. Krueger has been a Research Program Manager at the University of Wisconsin-Madison since 1995 and a founding member and Chief Executive Officer of Complete Phytochemical Solutions, LLC since 2010. He specializes in the development, validation and harmonization of methods for the authentication and quantification of bioactive compounds in dietary supplements, foods and beverages. Christian has published over 80 peer-reviewed manuscripts and holds numerous patents on natural product technologies. He is a member on several AOAC International Expert Review Panels, an Expert Reviewer with the AOAC Research Institute’s Performance Tested Method^SM Program, a member of the American Herbal Products Association’s Psychedelic Plants and Fungi Committee, a member of the American Botanical Council’s Botanical Adulterants Prevention Program and works closely with the United State Pharmacopeia (USP) to assist in development of monographs for the Dietary Supplement Compendia. He is an invited Advisory Council Member for Sonoran University of Health Sciences. His expertise in natural product chemistry, understanding of natural products supply chain management (growers, processors, formulators, and retail) and research experiences relating phytochemical structures to biologic function provides a unique skill set and background.

Andrew Birmingham

Andrew Birmingham is the Director of Analytic Service at Complete Phytochemical Solutions, LLC. He received his bachelor’s degree in biomedical biology from the University of Wisconsin-La Crosse and a master’s degree from the University of Wisconsin-Madison. While managing the laboratory, he is in the process of developing and maintaining a quality management system at Complete Phytochemical Solutions, LLC for ISO 17025:2017 accreditation.

Amy B. Howell

Amy B. Howell, PhD has been an Associate Research Scientist (now emeritus) at the Marucci Center for Blueberry Cranberry Research at Rutgers University in Chatsworth, NJ since 1993 and a founding member of Complete Phytochemical Solutions, LLC. She has been engaged in research that targets utilizing cranberry for prevention and management of bacterial diseases, including urinary tract infections (UTIs), stomach ulcers, and periodontal disease. Her primary research focus has been on isolating polyphenolic compounds from cranberry and determining their role in prevention of UTIs and collaborating on ex vivo clinical studies to determine effects of cranberry proanthocyanidins on uropathogenic bacterial adhesion in urine. She studies the pharmacokinetics and bioavailability of the structurally unique cranberry proanthocyanidins in an effort to determine site(s)of action and dose-response. Dr. Howell has been involved in method development for powdered cranberry supplements, working closely with regulatory teams from AOAC and USP (US Pharmacopeia) to determine standard methods for quantification of the bioactive compounds in cranberries.

Jess D. Reed

Jess Reed is Emeritus Professor of Nutrition at the University of Wisconsin-Madison with 40 years of postgraduate experience in research on the effects of dietary phytochemicals on the nutrition and health of animals and humans. He also directs the Reed Research Group (RRG) which consist of 3 core programs: Phytochemistry, Cardiovascular and Mucosal Immunity. The main thrust of our Phytochemistry Research Core is to determine the effects of oligomeric polyphenols (tannins) in foods on disease processes. This research effort includes development of phytochemical methods for characterization of structure and their interactions with proteins and polysaccharides, and mechanistic studies on the effects of tannins in cell culture and animal models of disease. We are currently studying the relationship between tannin structure and the ability of tannins to inhibit the adherence, invasion and colonization of mucosal epithelial cells by extra-intestinal pathogenic E. coli and other bacteria that infect the GI tract, the urogenital tract and mammary gland. Dr. Reed is also a founding member and Chief Science Officer of Complete Phytochemical Solutions, LLC (CPS) since 2010. CPS is a provider of contract research and analytical services for the food industry.