huseyin ayvaz ift 2013 poster - finall.pdf.pptx (read-only) · 2016-09-11 · title: huseyin ayvaz...
TRANSCRIPT
Applica(on of a Portable Infrared Technology (PIRT) for Screening Sugar Levels in Chipping Potatoes Huseyin Ayvaz1, Jennifer Moyseenko2, Ma6hew Kleinhenz2 and L. E. Rodriguez-‐Saona1
1 Department of Food Science and Technology, The Ohio State University, Columbus, OH, 43210 2 Ohio Agricultural Research and Development Center, The Ohio State University, Wooster, OH, 44691
Ø Sugars in potato tubers are key components influencing the quality of processed potato products. Maillard reacVon between reducing sugars (glucose and fructose) and free amino acids produces the color and flavor of the products [1]. AddiVonally, acrylamide which is a human neurotoxin and “probably carcinogen to humans” is formed through this reacVon between reducing sugars and mainly asparagine at high temperatures [2].
Ø Sucrose may also involve in Maillard reacVon through hydrolysis during frying. Reducing sugars can also form from sucrose via invertase enzyme acVvity during the storage. Regarding the Maillard reacVon, reducing sugars have been shown to be the limiVng factor. Several factors including genotype, the environmental condiVons during growth and post-‐harvest factors such as storage can affect the sugar levels in potatoes [3].
Ø The most popular methods used to determine sugar and asparagine levels in raw potatoes requires intensive sample preperaVons and chromotographic separaVons such as HPLC or GC, which are Vme consuming and labor intensive, making efficient assays for rapid, accurate and “in-‐field” tesVng desirable.
Ø Infrared spectroscopy has been used for numerous food applicaVons related to quality control and safety of our food supply providing valuable informaVon about the biochemical composiVon of the samples, especially in the fingerprint region. Ø MiniaturizaVon of vibraVonal spectroscopy components has allowed the development of portable and handheld systems that offer simplicity, speed, selecVvity and performance similar to that of benchtop instruments used in laboratories [4]. Ø In order to compare the infrared models, raVo of predicVon to validaVon (RPD) can be used. It is calculated by dividing standard deviaVon in validaVon set samples to standard
error of predicVon (SEP) in validaVon set. RPD values between 2.5 and 3.0 is considered as good and above RPD value of 3.0, the predicVon is classified as excellent [5].
Among many compounds present in potato tuber, sugars are considered as the most important ones influencing the quality of processed potato products. Maillard reacVon between reducing sugars and free amino acids contributes to the color and flavor of the products, remarkably. AddiVonally, acrylamide, which is a human neurotoxin and a “probable carcinogen to humans”, is formed primarily through this reacVon between reducing sugars and asparagine at high temperatures. Currently, the most common methods used to determine sugars and asparagine in raw potatoes require intensive sample preperaVon and chromotographic separaVon which are Vme consuming and labor intensive. Efficient assays for rapid, accurate and “in-‐field” tesVng are thus desirable. Our main objecVve was to evaluate the feasibility of using a portable infrared technology instrument (PIRT) for the rapid quanVtaVon of sugars and asparagine in chipping potato varieVes. Samples represenVng a total of 47 experimental clones or commercial varieVes were obtained from the Ohio Agricultural Research and Development Center in Wooster (OH). Sugar levels were determined using HPLC-‐RID as reference method and found to encompass a wide range of levels: 0.8-‐6.9 mg glucose, 0.5-‐9.2 mg fructose and 1-‐4.7 mg sucrose per 1 g fresh weight. Asparagine levels (4.9-‐23.8 mmol/kg fresh weight) were determined using GC-‐FID as reference method. Ten variety of potatoes were also made into slices and fried for acrylamide analysis by LC-‐MS/MS.
Soluble solids obtained from raw potatoes following very simple sample preperaVons (centrifugaVon or mixing with water) were used for infrared spectra collecVons. The potatoes’ sugar and asparagine levels were correlated with the infrared spectra and the resulVng ParVal Least Squares Regression models (n=37) were used to predict sugar and asparagine levels of unknown potato samples (n=11). Excellent linear correlaVons between predicted and reference values from HPLC and GC were obtained for the independent sample set. PLSR models gave high rPred (correlaVon coefficient for predicVon, >0.90) and very low standard errors of predicVons (SEP). RaVos of predicVon to validaVon (RPD) ranged between 2.2 and 5.0 indicaVng reliable and robust models for the predicVon of sugar and asparagine levels in potatoes.
In contrast to chromotographic methods, PIRT allows for the rapid, inexpensive, high throughput and accurate measurement of tuber sugar and asparagine levels in-‐field and in storage. PIRT also shows promise for the predicVon of acrylamide levels in fried potato chips. Therefore, PIRT can significantly benefit potato breeding as well as certain aspects of crop management, producVon and research.
RESULTS
REFERENCES 1. Rodriguez-‐Saona, L. E., & Wrolstad, R. E. (1997). Influence of Potato ComposiVon on Chip Color Quality. American Journal of Potato Research, 74, 87-‐106. 2. Friedman, M. (2003). Chemistry, biochemistry, and safety of acrylamide. A review. Journal of Agricultural and Food Chemistry, 51, 4504-‐4526. 3. Kumar, D., Singh, B. P., & Kumar, P. (2004). An overview of the factors affecVng sugar content of potatoes. Annals of Applied Biology, 145, 247-‐256. 4. Ellis, D.I., & Goodacre, R. (2006). Metabolic fingerprinVng in disease diagnosis: biomedical applicaVons of infrared and Raman spectroscopy. The Analyst, 131, 875-‐885. 5. Saeys, W., Mouazen, A.M., & Ramon, H. (2005). PotenVal for Onsite and Online Analysis of Pig Manure using Visible and Near Infrared Reflectance Spectroscopy. Biosystems Engineering, 91, 393-‐402
INTRODUCTION
ABSTRACT
Ø To develop simple, sensiVve and robust methods for quanVficaVon of sugars and asparagine levels in raw potatoes using a portable Cary 630 FTIR system based on highly specific MIR spectroscopic signature profiles in combinaVon with supervised pa6ern recogniVon techniques
Ø Evaluate the performance of different accessories for infrared spectra collecVon (transmission cell, single bounce ATR and five bounce-‐temperature controlled ATR) Ø Evaluate the spectra from raw potatoes for predicVng acrylamide formaVon upon frying
OBJECTIVES
MATERIALS AND METHODS
Ø Using the same spectra, separate sucrose, glucose, fructose, reducing sugars and asparagine PLSR models were successfully developed (rPed > 0.9) Ø MulVvariate models accurately predicted sugar levels in independent sample sets with RPD value being higher than 2.5 for most of the models, which indicates that models
can be used for quanVtaVve measurements. Only excepVon was fructose models whose RPD values ranged between 2.2 and 2.4. It may be be6er to use total reducing sugar model unless individual fructose result is needed.
Ø Asparagine models gave very high RPD values (between 3.4 and 3.9) Ø Using the same infrared regions as sugar models from raw potato spectra, acrylamide formed upon deep frying of 10 varieVes of potatoes was accurately predicted. However,
acrylamide levels found in fried chips were remarkably higher than those of most of the industrial potato chips. SVll, the preliminary data is very promising for future research. Ø For all the IR sample preparaVon, 3000 and 12000 g centrifugaVon was applied for potatoes to obtain supernatant and develop the models. Since 3000 g provided similarly
good models, there is no need to do 12000 g, which may also increase the applicability of the propose techniques (Only results for 3000 g are shown). Ø Different accessories of Cary 630 FTIR used performed similarly, which indicates that any of them can be chosen for the applicaVons (transmission cell, one bounce and five
bounce ATR`s) Ø Using the portable systems is simple, cost-‐effecVve and requires low sample volume; once the instrument is purchased, there are minimal operaVonal costs involved on
performing the tests. AddiVonally, portable systems provide increased flexibility and great potenVal for in-‐field applicaVons compared to bench-‐top IR systems or chromotographic systems such as HPLC and GC, which can only be used in a laboratory setng.
DISCUSSION & CONCLUSIONS
Table 1 PredicVon Performance Summary of PLSR models developed for sugars in potatoes using different accessories of portable Cary 630 IR systems Analyte Range
(mg/g) Technique Factors rVal SECV rPred SEP RPD
Sucrose
Transmission cell 8 0.97 0.27 0.97 0.32 3.6 1.0 -‐ 4.7 Single bounce ATR 5 0.98 0.26 0.98 0.24 5.0
5 bounce ATR 5 0.97 0.26 0.97 0.32 3.7
Glucose Transmission cell 6 0.94 0.49 0.98 0.51 3.4
0.8 – 6.9 Single bounce ATR 6 0.95 0.47 0.97 0.51 3.1 5 bounce ATR 3 0.92 0.52 0.98 0.62 2.6
Fructose
Transmission cell 8 0.94 0.72 0.96 0.87 2.4 0.5 – 9.2 Single bounce ATR 8 0.93 0.81 0.91 0.87 2.2
5 bounce ATR 7 0.94 0.75 0.90 0.82 2.3 Glucose
+ Fructose
Transmission cell 7 0.95 1.04 0.98 1.28 2.6 1.3 – 15.5 Single bounce ATR 8 0.94 1.11 0.97 1.20 3.1
5 bounce ATR 7 0.94 1.25 0.92 1.46 2.4
Table 2 PredicVon Performance Summary of PLSR models developed for asparagine in potatoes using different accessories of portable Cary 630 IR systems
Analyte Range (mmol/kg)
Technique Factors rVal SECV rPred SEP RPD
Asparagine
Transmission cell 8 0.96 0.96 0.97 1.16 3.9 4.9 – 23.8 Single bounce ATR 8 0.96 0.94 0.96 1.30 3.4
5 bounce ATR 6 0.97 1.09 0.97 0.87 3.8
3
7
11
15
19
23
3 7 11 15 19 23
Pred
icted aspa
ragine
con
centra(o
n (m
mol/kg)
Asparagine concentra(on measured by GC-‐FID (mmol/kg)
ASPARAGINE TRANSMISSION
3
7
11
15
19
23
3 7 11 15 19 23
Pred
icted aspa
ragine
con
centra(o
n (m
mol/kg)
Asparagine concentra(on measured by GC-‐FID (mmol/kg)
ASPARAGINE 5 BOUNCE ATR
1
6
11
16
21
1 6 11 16 21 Pred
icted acrylamide concen
tra(
ons (mg/
kg)
Acrylamide concentra(on measured by LC-‐MS/MS (mg/kg)
ACRYLAMIDE TRANSMISSION
Factor : 8 SECV : 0.3 mg/kg rVal : 0.99
0.25
1.25
2.25
3.25
4.25
5.25
6.25
0.25 1.25 2.25 3.25 4.25 5.25 6.25
Pred
icted sucrose concen
tra(
on (m
g/g)
Sucrose concentra(on measured by HPLC-‐RID (mg/g)
SUCROSE TRANSMISSION
0.5
2.5
4.5
6.5
8.5
10.5
12.5
14.5
16.5
1 3 5 7 9 11 13 15 17
Pred
icted redu
cing su
gar con
centra(o
n (m
g/g)
Reducing sugar concentra(on measured by HPLC -‐RID (mg/g)
REDUCING SUGARS TRANSMISSION
0.25
1.25
2.25
3.25
4.25
5.25
6.25
0.25 1.25 2.25 3.25 4.25 5.25 6.25
Pred
icted sucrose concen
tra(
on (m
g/g)
Sucrose concentra(on measured by HPLC-‐RID (mg/g)
SUCROSE 1 BOUNCE ATR
0.5
2.5
4.5
6.5
8.5
10.5
12.5
14.5
16.5
0.5 2.5 4.5 6.5 8.5 10.5 12.5 14.5 16.5
Pred
icted redu
cing su
gar con
centra(o
n (m
g/g)
Reducing sugar concentra(on measured by HPLC-‐RID (mg/g)
REDUCING SUGARS 1 BOUNCE ATR
0.25
1.25
2.25
3.25
4.25
5.25
6.25
0.25 1.25 2.25 3.25 4.25 5.25 6.25
Pred
icted sucrose concen
tra(
on (m
g/g)
Sucrose concentra(on measured by HPLC-‐RID (mg/g)
SUCROSE 5 BOUNCE ATR
0.5
2.5
4.5
6.5
8.5
10.5
12.5
14.5
16.5
0.5 2.5 4.5 6.5 8.5 10.5 12.5 14.5 16.5
Pred
icted redu
cing su
gar con
centra(o
n (m
g/g)
Reducing sugar concentra(on measured by HPLC-‐RID (mg/g)
REDUCING SUGARS 5 BOUNCE ATR
Figure 2. PLSR calibration and validation plots for asparagine 1
( , represent samples in calibration and validation groups, respectively) 2
Figure 1. PLSR calibration and validation plots for sugars 1 ( , represent samples in calibration and validation groups, respectively) 2
PLSR Modeling ü Spectral transformation
Normalize 2nd derivative (25)
ü Selected spectral region 900-1500 cm-1 for sugars 900-1700 cm-1 for asparagine
ü 37 varieties used in calibration and 11 varieties used in validation set
Figure 3. Average of all spectra collected using transmission and five bounce 1
temperature controlled unit of portable Cary 630 IR system 2 PLSR Modeling
ü Supernatant of 1 g raw potato powder
ü Spectral transformation 2nd derivative (35)
ü Selected spectral region 900-1500 cm-1 for acrylamide
ü Only 9 varieties fried and used as preliminary
Figure 4. PLSR plot for acrylamide content of potato chips made 1
Sugar extracVons (80 % ethanol)
Amino acid extracVon (EZFaast kit)
Asparagine analysis using GC-‐Flame IonizaVon Detector
Glucose, fructose and sucrose measured by
HPLC-‐RID
Blend with liquid Nitrogen
Acrylamide extracVon and
analysis using LC-‐MS/MS
Deep frying (180 0C, 2.5 min)
10 µL supernatant on transmission cell (30 µm path
length)
1 µL supernatant on single bounce ATR
(1 min vacuum drying)
1 g potato centrifuged at 3000 g for 15 minutes
2 g potato + 25 ml water sVrred for 15 minutes
75 µL supernatant on 5 bounce ATR (2.5 min vacuum drying at 65 0C) PLSR
900140019002400290034003900
SNV(abs)
Wavenumber (cm-‐1)
5 Bounce ATR Transmission