Multi-pulsed high hydrostatic pressure treatment for inactivation and injury of Escherichia coli

Escherichia coli cells in peptone water were pressurized at 300?MPa at ambient temperature with no holding time (pulse series) and with a total holding duration of 300?s for single- (300?s?×?1 pulse) and multi-pulsed (150?s?×?2 pulses, 100?s?×?3 pulses, 75?s?×?4 pulses, 60?s?×?5 pulses, 50?s?×?6 pulses and 30?s?×?10 pulses) high hydrostatic pressure (HHP) treatments. Multi-pulsed HHP treatment with no holding time indicated that as the pulse number increased the number of inactivated and injured cells also increased. Holding time had significant effect on the inactivation of E. coli. There was low inactivation difference between single- and multi-pulsed HHP treatments with holding time. Escherichia coli cells showed at least 1.6 log₁₀ more reduction on selective medium than the non-selective medium indicating that more than 95?% of the survivors severely injured for both single- and multi-pulsed treatments with holding time. Although the inactivation difference was low between single- and multi-pulsed HHP treatments, storage at 4?°C revealed that there was less recovery from injury for multi-pulsed HHP treatment.     

Destruction of Salmonella enteritidis inoculated in liquid whole egg by high hydrostatic pressure: comparative study in selective and non-selective media

The destruction of Salmonella enteritidis inoculated in liquid whole egg at approximately 10⁷−10⁸cfu ml^−lwas studied under combinations of pressure (350 and 450 MPa), temperature (50, 20, 2 and −15°C) and time (5, 10, 15 min and cycles of 5+5 and 5+5+5 min). One non-selective medium (tryptone soy agar) and two selective media (brilliant green agar and salmonella-shigella) were used to evaluate viability of S. enteritidis after pressurization. The inactivation rate increased with pressure and exposure time, being minimal at 350 MPa and −15°C for 5 min (over 1 log₁₀of reduction) and reaching total inactivation (8 log₁₀of reduction) in several treatments at 50°C. Treatments in cycles showed greater effectiveness than continuous treatments of the same total time. The effect of pressure was enhanced by elevated temperatures. The higher counts were obtained in the non-selective medium, indicating the presence of injured cells after pressure treatment. D -values obtained for two temperatures (2 and 20°C) and different times (0–60 min) under controlled pressure (400 MPa) showed that microbial inactivation followed a first-order kinetics with a decimal reduction time evaluated in tryptone soy agar medium of 9·5 min at 2°C and 8·8 min at 20°C.     

Recovery of Salmonella from refrigerated preenrichment and refrigerated enrichment media

The productivity of the standard cultural procedure for the isolation of Salmonella using preenrichment in buffered peptone water (BPW) followed by inoculation into Rappaport-Vassiliadis (RV), tetrathionate (TBG) (Difco) and selenite-cystine (SC) (Difco) enrichment broths, was compared with that using preenrichment and enrichment cultures which had been held under refrigeration (72 h at 5–10°C). Seventy-seven of 251 samples of food products were found to contain Salmonella. Refrigerated preenrichment cultures inoculated into RV medium (43°C), TBG broth (43°C) and SC broth (37°C) yielded salmonellae from 93.5, 85.7 and 54.5% contaminated samples, respectively. Refrigerated cultures in RV, TBG and SC broths, inoculated onto three selective plating media, identified 100, 87.0 and 41.6% of the contaminated samples, respectively.The selective plating medium brilliant green deoxycholate agar was at least as productive as brilliant green sulpha agar and bismuth sulphite agar, when streaked from RV and TBG broths, but was less effective when streaked from SC broth.     

Evaluation of Thin Agar Layer Method for Recovery of Heat-Injured Foodborne Pathogens

ABSTRACT The Thin Agar Layer (TAL) method of Kang and Fung 1998 was used to enumerate heat‐injured (55 °C for 10 min) foodborne pathogens. This method involves overlaying 14 mL of nonselective medium (Tryptic Soy Agar, TSA) onto a prepoured, pathogen‐specific, selective medium. The recovery rate with the TAL method was compared with the rate for TSA and pathogen‐specific, selective media. No significant differences occurred between TSA and TAL (P > 0.05) for enumeration of heat‐injured Escherichia coli O157:H7, Listeria monocytogenes, Salmonella typhimurium, Staphylococcus aureus, and Yersinia enterocolitica, and both recovered significantly higher numbers than selective media (P < 0.05). The TAL method is a 1‐step, convenient procedure for recovery of heat‐injured cells.     

Thermal inactivation of Salmonella spp. in ground chicken breast or thigh meat

Thermal inactivation of a four‐strain mixture of Salmonella spp. was determined in chicken breast and thigh meat. Inoculated meat was packaged in bags and then completely immersed in a circulating water bath and held at 55, 57.5, 60 and 62.5 °C for pre‐determined lengths of time. When the surviving bacteria were enumerated on tryptic soya agar supplemented with 0.6% yeast extract and 1% pyruvate (non‐selective medium), D‐values (time to inactivate 90% of bacteria) in chicken breast meat were 6.08, 4.77, 3.00 and 0.66 min at 55, 57.5, 60 and 62.5 °C, respectively; the values in thigh meat ranged from 11.48 min at 55 °C to 0.84 min at 62.5 °C. As expected, the measured heat resistance was lower when the recovery medium was selective (xylose lysine deoxycholate agar). Thermal death time values from this study will assist food processors in designing the HACCP plans to effectively eliminate Salmonella spp. in cooked chicken breast and thigh meat.     

Short communication: Evaluating the recovery potential of injured cells of Listeria innocua under product temperature-abuse conditions and passage through simulated gastrointestinal fluids

Ice cream handling and serving conditions on the consumer side may result in temperature abuse before consumption. Under some extreme conditions, even the sporadic presence of injured bacterial cells might pose a health risk due to the possibility of recovery of those cells. We conducted this investigation to evaluate the potential of injured cells of Listeria innocua to recover under ice cream temperature abuse conditions and on exposure to simulated gastrointestinal (GI) fluids. Ice cream mix samples (42% total solids), spiked with 4 log₁₀ cfu/g of Listeria innocua, were thermally treated at 69°C for 30 min. Potential heat-injured cells were recovered in buffered Listeria broth (BLEB), followed by isolation on Listeria-specific modified Oxford agar (MOX). The ice cream mix samples, containing potentially injured cells of Listeria innocua, were followed through overnight aging (7°C), freezing (?3.3°C), and overnight hardening (?40°C) steps to obtain the final ice cream samples. To simulate temperature abuse conditions, the samples were held for 12 h at 4.4°C, followed by 30 min at room temperature (22°C); this treatment was considered the first cycle of temperature abuse. To generate a worst-case scenario, the samples were exposed to 3 such consecutive temperature abuse cycles. At the end of each cycle, direct plating was done on MOX to recover viable cells, and BLEB enrichment verified the presence of potential injured cells. In addition, the ice cream samples, containing potential injured cells, were passed through simulated GI fluids. As a first step, samples were mixed (1:1) with simulated gastric fluids (pH 1.0 and 2.0 before mixing) and held at 37°C in a shaker incubator. Samples drawn at 15, 30, and 60 min were analyzed for viable and potential injured cells. To study the effect of sequential transit through simulated intestinal fluid, a mixture of ice cream and gastric fluid (1:1) from the gastric fluid experiment above was added to simulated intestinal fluid (pH 6.8) and held at 37°C. Samples were analyzed at 30 and 360 min for viable and potential injured cells. Three trials were conducted and the samples collected in duplicates. The temperature abuse or GI fluid exposure studies did not result in the recovery of potential injured cells of Listeria innocua in the ice cream samples under the conditions tested. Exposure to gastric fluids, however, did not eliminate the potential injured cells. Further studies are necessary to understand the exact risk implications of these findings.     

Evaluation of Preston medium for detection ofCampylobacter jejuni in vitroand in artificially and naturally contaminated poultry products

A short pre-enrichment time (4 h at 37°C) allowed the recovery of low levels ofCampylo- bacter jejuniamong an accompanying flora. Incubation time in a non-selective medium should not be further elongated because of outgrowth of the accompanying flora suppressing development of low numbers ofC. jejuni. Immediate selective enrichment (24 h at 42°C in Preston medium) also has the ability for detection of low levels of the pathogen bothin vitro(10–1000 cfu 100 ml⁻¹) and in food samples (<10–1000 cfu 10 g⁻¹). Exclusion of a prior non-selective enrichment step rendered this last procedure less laborious. Application of a 24 h selective enrichment in Preston medium at 42°C for detection ofC. jejuniin poultry products identified 40% of the samples (n=60) to be contaminated with the pathogen. Elongation of incubation time to 48 h with or without blood supplementation to the Preston medium did not significantly improve detection of the pathogen. Supplementation of ferrous sulphate, sodium metabisulphite and sodium pyruvate, each at 0.25 g l⁻¹and microaerophilic atmosphere fulfilled the requirement of enhancing the aerotolerance ofC. jejuniand allowed recovery of the pathogen in 24 h.     

Heating and Storage Conditions Affect Survival and Recovery of Listeria monocytogenes in Ground Pork

The effect of heating rate, heating atmosphere, and meat age on survival of Listeria monocytogenes was examined in ground pork, as well as the ability of injured cells to recover during storage under air and vacuum packaging at 4 and 30°C. Significantly more survivors were observed when samples were heated at 1.3°C/min than at 8.0°C/ min. Cells inoculated into 3-month-old pork were more sensitive to heating than cells inoculated into fresh ground pork (D_62°C= 5.2 min and 7.7 min, respectively). More survivors were detected when the meat was heated aerobically than anaerobically. However, storage under vacuum at 4 or 30°C resulted in faster recovery compared with cells packaged in air. Thus, heating and packaging conditions affected ability of this pathogen to survive and recover after a heat process in pork.     

Listeria cross contamination levels in raw ice cream mix can serve as a predictor of their potential presence as heat-injured cells

Listeriosis is a life-threatening infection caused by foods contaminated with Listeria monocytogenes. Some of the major ice cream recalls in recent years reaffirm the ability of this food-borne pathogen to survive in diverse dairy processing environments and cause cross contamination. Inspection reports revealed certain lapses in implementing adequate hygienic practices for Listeria persistence in the processing environment, leading to cross contamination of ice cream. The higher levels of cross contamination of raw ice cream mix might result in random heat-injured cells when exposed to minimum heat treatment (69°C for 30 min). These heat-injured cells could later recover under abusive storage and handling conditions and pose a health risk. Evidence about the presence of injured cells in ice cream mix may thus prove useful to establish the overall Listeria risk, which was the aim of this study. Challenge studies were conducted to evaluate the dose-dependent presence of heat-injured cells of Listeria. Ice cream mix formulations of 4 different types (36, 40, 42, and 45% total solids) were inoculated at 2.0, 3.0, and 4.0 log cfu/g levels of Listeria innocua (an established surrogate). The dose levels were selected based on a likely cross contamination on the raw side from environmental Listeria, especially due to their resident nature and growth in harborage sites. The samples were exposed to minimum heat treatment (69°C for 30 min) and the survivors, including heat-injured cells, were enumerated using standard protocols. A binary logistic regression model was fitted for evaluating the severity of risk. The influence of total solids, water activity, and pH variability were also studied on Listeria survival. The enrichment protocol, using buffered Listeria enrichment broth, followed by plating on modified oxford agar and Rapid L'mono medium, revealed the random presence of heat-injured cells in buffered Listeria enrichment broth, only at the highest dose level of 4+ logs. Any potential risk from heat-injured cells was thus limited only to the highest levels of cross contamination, irrespective of the type of the mix. Significantly, none of the pasteurized ice cream mix samples supported the recovery of any heat-injured cells of Listeria during 72 h holding at 7°C, even at the highest dose level of 4+ logs, under the conditions of experimentation. The level of cross contamination (dose) emerged as a predictor of the potential presence of heat-injured cells of Listeria exposed to minimum pasteurization treatment.     

Flow property of globulin from red bean (Phaseolus angularis)

Red bean globulin (RBG) dispersions exhibited pseudoplastic behavior, and the flow data were fitted to the power-law model. With increasing protein concentration, there were progressive increases in consistency coefficient (m), apparent viscosity and yield stress, suggesting protein–protein interactions. The flow behavior index (n) was decreased indicating enhanced pseudoplasticity. Ionic strength affected the flow behavior of RBG, with a reversal in its effect at a specific salt concentration (0.5–1.0 M). Addition of chaotropic salts and protein structure perturbants (urea, sodium dodecyl sulfate and dithiothreitol) caused changes in flow behavior, probably due to dissociation of the oligomers and protein unfolding. Pre-heat treatments at a temperature of 80°C, above the onset temperature of RBG, caused dramatic decreases in n and increases in m and apparent viscosity suggesting protein denaturation/aggregation. When RBG dispersions were heated from 25 to 90°C at a fixed rate (1.7°C/min), there was an initial decrease in apparent viscosity, probably due to a lowering of medium viscosity. The viscosity started to increase at around 78°C, the onset temperature of RBG; and sharp increases were observed from 80 to 90°C. Changes in apparent viscosity with increasing temperature (25–60°C) obeyed the Arrhenius law, with similar E_a values at three different shear rate values, suggesting little temperature dependence of viscosity at lower temperature range.     

High-Pressure Destruction Kinetics of E. coli(29055) in Apple Juice

High‐pressure (HP) destruction of Escherichia coli (29055) in apple juice was evaluated (150 to 400 MPa; 0 to 80 min at 25 °C) based on a dual destruction behavior consisting of pressure pulse and pressure hold (first order rate) effects. Enumeration was carried out in brain‐heart infusion agar (BHIA) and violet‐red bile agar (VRBA) to differentiate between surviving cells with and without injury. A pressure pulse at 400 MPa destroyed the entire population (10⁸ CFU/mL) of E. coli. During pressure‐hold, D values (decimal reduction time) of E. coli decreased with an increase in pressure, and D values from BHIA (survivors including injured cells) were higher than from VRBA (survivors excluding injured cells). Thus, an increasing number of cells were initially injured than killed with HP treatment until finally killed. Pressure dependency of D was described by z‐value model with associated z_p values (pressure range to result in a decimal change in D values) of 126 and 140 MPa on BHIA and VRBA.     

Comparison of Modified Direct-Plating Procedures for Recovery of Injured Escherichia coli

E. coli ATCC 11775 was heat injured at 59, 62, and 69°C, and recovered by a modification of the conventional coliform count on VRB. Four enrichment media, four selective overlay media, and four preincubation periods were tested to determine the optimal method for recovery of injured coliforms. Preincubation of injured E. coli for 2 hr on tryptone glucose extract (TGE) medium or plate count agar (PCA) significantly increased recovery of injured cells. Single strength violet red bile (VRB) agar gave higher recoveries than double strength VRB. No significant differences were observed between counts on boiled and sterilized VRB. We recommend the use of sterilized VRB since it is more convenient to prepare and store. Of the four enrichment and selective media examined, the recovery media of choice were TGE or PCA when used in combination with the sterilized VRB as the overlay medium. Therefore, preincubation for 2 hr with TGE or PCA overlayed with sterilized VRB is recommended for optimal recovery of injured E. coli.     

Multi-pulsed high hydrostatic pressure treatment for inactivation and injury of Escherichia coli

Escherichia coli cells in peptone water were pressurized at 300 MPa at ambient temperature with no holding time (pulse series) and with a total holding duration of 300 s for single- (300 s × 1 pulse) and multi-pulsed (150 s × 2 pulses, 100 s × 3 pulses, 75 s × 4 pulses, 60 s × 5 pulses, 50 s × 6 pulses and 30 s × 10 pulses) high hydrostatic pressure (HHP) treatments. Multi-pulsed HHP treatment with no holding time indicated that as the pulse number increased the number of inactivated and injured cells also increased. Holding time had significant effect on the inactivation of E. coli. There was low inactivation difference between single- and multi-pulsed HHP treatments with holding time. Escherichia coli cells showed at least 1.6 log₁₀ more reduction on selective medium than the non-selective medium indicating that more than 95 % of the survivors severely injured for both single- and multi-pulsed treatments with holding time. Although the inactivation difference was low between single- and multi-pulsed HHP treatments, storage at 4 °C revealed that there was less recovery from injury for multi-pulsed HHP treatment.     

Influence of substrate and low temperature on growth and survival of verotoxigenicEscherichia coli

Growth and survival of verotoxigenicEscherichia coli(VTEC) were studied in tryptic soy broth (TSB), brain–heart infusion (BHI), and whole milk at 12, 9.5, 8.5, 7.5, 6.5, and 5.5°C (±0.2°C). Colony forming units (cfu) were enumerated by surface plating on eosin methylene blue (EMB) agar plates and incubating at 37°C for 24 h. Growth curves were plotted and lag and generation times were determined. Results indicated that the average generation times for each strain was similar in each medium (4.6±0.9 h) at 12°C. However, at 9.5°C the average generation time differed significantly among the media (P<0.05). The average generation time was longer in BHI compared to TSB and whole milk at 9.5°C (14 h in BHI vs 11 h in milk and TSB). Also, the generation time of individual strains in BHI varied from 9.5–22 h at 9.5°C. Decrease of temperature to 8.5°C resulted in lack of growth of two strains in BHI. At 7.5°C none of the strains grew in BHI, one grew in TSB, whereas all grew in milk. A further decrease of temperature to 6.5°C did not allow growth of any of the strains in TSB or BHI whereas all grew in milk. Addition of 5% lactose to TSB and BHI enabled survival and growth of all strains at 6.5°C. The results of this study demonstrate the influence of growth medium composition on the minimum growth temperature of verotoxigenicE. coli.     

Prediction of Injured and Dead Inactivated Escherichia coli O157:H7 Cells after Heat and Pulsed Electric Field Treatment with Attenuated Total Reflectance Infrared Microspectroscopy Combined with Multivariate Analysis Technique

Attenuated total reflectance infrared microspectroscopy combined with soft independent modeling of class analogy (SIMCA) multivariate technique was used to differentiate between alive, dead, and injured Escherichia coli O157:H7 cells. E. coli O157:H7 cells were suspended in citrate phosphate buffer at pH 4.0 and 7.0 and treated by heat at 54 °C for 5, 10, 20, and 90 min or pulsed electric field (PEF) at 35 kV/cm for 10, 25, 50, and 60 pulses. The SIMCA analysis confirmed that major contribution to the discrimination of the untreated and treated E. coli cells were (1) the amide I band at 1,638 and 1,618 cm^?1 corresponding to changes in β-pleated secondary protein structure (heat-treated cells at pH 7.0 and 4.0, and PEF-treated cells at pH 4.0), and (2) the bands at 1,078 and 993 cm^?1 corresponding to changes in P?=?O (PO ₂ ^? ) stretching of phosphodiesters or lipopolysaccharides and C–O–C or C–O stretching of different polysaccharides (PEF-treated cells at pH 7.0). The use of partial least squares regression analysis allowed for correctly predicting the survivors of the thermal treatment. Injured cells could be estimated from the comparison of cell counts predicted in nonselective and selective plating media with sodium chloride and bile salts. The prediction results yielded inactivation values with a coefficient of determination (R ²) of 0.83 or higher and a standard error of cross validation between 0.11 and 0.37 log cycles of inactivation.     

Bacillus cereus emetic toxin production in cooked rice

Three mesophilic strains of Bacillus cereus known to produce emetic toxin were used to model germination, growth and emetic toxin production in boiled rice cultures at incubation temperatures ranging from 8°C to 30°C. Minimum temperatures for germination and growth in boiled rice were found to be 15°C for all strains. Toxin production at 15°C was found to be significantly greater (P<0·01; reciprocal toxin titre of 373±124) than at 20°C and 30°C (reciprocal toxin titres 112±37 and 123±41, respectively). Toxin production became detectable after 48 h incubation at 15°C, with a maximum titre reached by 96 h. At 20°C and 30°C, toxin production was detected at 24 h incubation, with a maximum titre reached by 72 h. Toxin production at 15°C was detectable at lower bacterial counts (6·2 log₁₀ cfu g⁻¹), than with incubation at 20°C and 30°C (>7·0 log₁₀ cfu g⁻¹). In this study, the lower temperature limit for germination and growth on solid laboratory medium was found to be 12°C for all strains, i.e. 3°C lower than that observed in boiled rice.     

Survival of Escherichia coli O157:H7 in Apple Cider as Affected by Dimethyl Dicarbonate, Sodium Bisulfite, and Sodium Benzoate

Survival of Escherichia coli O157:H7 in apple cider containing no preservatives, 0.025% dimethyl dicarbonate (DMDC), 0.045% sodium benzoate (SB), 0.0046% sodium bisulfite (NaS; 65.5% sulfur dioxide), or a combination of NaS and SB (NaS/SB) and stored at 4, 10, and 25°C was evaluated. E. coli O157:H7 survived for up to 18 days at 4,10, and 25°C in unpreserved apple cider. At 4 and 10°C, DMDC was most efficient at inactivating E. coli O157:H7, generally followed by NaS/SB SB, and NaS (p<0.05). E coli O157:H7 was more resistant to preservatives at 4°C than at 25°C (P < 0.05). E. coli O157:H7 was sublethally injured in cider containing preservatives, and to a lesser extent, in unpreserved cider. Generally, injury was more pronounced in cider containing DMDC, followed by NaS/SB, SB, and NaS (p<0.05).     

Inactivation of pathogenic Escherichia coli in skimmed milk using hhigh hydrostatic pressure

The pressure resistance of a range of pathogenic Escherichia coli strains was determined in skimmed milk. Pressure resistances varied widely among the 12 strains tested, with the two most resistant being an enterotoxigenic E. coli (NCTC 11601) and an enteroinvasive E. coli (NCTC 9706). A pressure treatment of 500 MPa for 40 min gave only a 4-log reduction of these two most resistant strains, but after a treatment of 600 MPa for 30 min, no survivors of either strain could be detected (>7 log reduction). The effect of the length of time in the stationary phase on the pressure resistances of these two strains was determined. Pressure resistance (600 MPa for 10 min at 20°C) was determined after 12 h incubation at 37°C in tryptone soya broth with 0.6% yeast extract added (TSBYE) (early stationary phase) and at 24-h intervals after this. The pressure resistances of these strains did not change significantly for the first 5 days after the onset of the stationary phase, after which time their resistances decreased. When cells were pressure-treated 7 days after they went into stationary phase, no survivors could be detected.     

A simple procedure for the detection of small numbers of enterotoxigenic Clostridium perfringens in frozen meat and cod paste

By use of commercially available pouches and dehydrated modified Handford medium, a simple method for the selective detection of small numbers of Clostridium perfringens spores in frozen meat and cod paste has been developed. By a direct method, 1 × 10²/g or more spores were detectable and after enrichment culture, the sensitivity of the direct method was increased by 10 to 100 times. One hundred and thirty of 319 samples (40.8%) of frozen meat and cod fish paste proved positive for presumptive C. perfringens by the enrichment method. The incidences of presumptive C. perfringens in frozen mutton (50.0%), rabbit meat (65.0%), cod paste (50.0%) and horse meat (42.4%) were relatively higher than those in frozen beef (25.0%) and pork (23.8%). By the direct method, the incidence decreased to 3.1% (10/319). Two enterotoxigenic strains were found among 100 verified C. perfringens isolates from 58 samples.     

Characterization of starch and cell walls from mature-green ambarella (Spondias cytherea Sonnerat) and their enzymatic hydrolysis

Juice from mature-green ambarella contains starch, a characteristic detrimental to its visual appearance due to the white sediment formed upon storage. The purpose of this work was to evaluate the effects of starch and cell wall degrading enzymes on juice residual starch and content in soluble sugars. Starch and cell walls from mature-green ambarella fruits were purified and characterized. Starch was found to contain 21.0% amylose, 78.1% amylopectin and 0.9% other minors compounds. Its average granule size was 20 μm. Its thermal characteristics were: temperatures of onset (T _o = 57.8 °C), peak (T _p = 65.6 °C), and conclusion (T _c = 72.6 °C) of gelatinization and the endothermic enthalpy (ΔH _{gelatinisation} = 12.4 J g^?1). Cell walls represented 2.8% of the edible fresh matter and were mainly constituted of highly methylated (HM) pectic substances and cellulose. The amylolytic preparations we tested, AMG^® 300 L and Hazyme^® C, showed similar behaviours in terms of starch hydrolysis and profit of Brix degree obtained. With 200 μg g^?1 of AMG^® 300 L or Hazyme^® C, the degree of amylolysis of coarse ambarella puree was close to 50% and its increased up to 70% with enzymes concentrations higher than 1,000 μg g^?1 (gelatinization at 75 ± 5 °C for 15 min followed by starch amylolysis at 60 ± 5 °C for 15 min). Total hydrolysis of ambarella starch is possible when pectinolysis occurred before amylolysis treatment, probably because of the fluidification of the medium by the pectocellulolytic enzymes. Pectinex^® Ultra SP-L was the most efficient preparation to degrade the ambarella pectins (～80% of cell wall uronides liberated from 120 mg g^?1 of purified cell walls within 1 h at 30 °C, pH 2.7).