Characterization by Size-Exclusion Chromatography with Refractive

11 Characterization by Size-Exclusion Chromatography with Refractive Index and Viscometry Downloaded by UNIV OF ALABAMA on November 21, 2016 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/ba-1995-0247.ch011

Cellulose, Starch, and Plant Cell Wall Polymers Judy D. Timpa† Agriculture Research Service, Southern Regional Research Center, U.S. Department of Agriculture, New Orleans, LA 70179

High molecular weight natural polymers are difficult to characterize because isolation, extraction, and dissolution often degrade the polymers and no good molecular weight standards exist. In our laboratory, cellulose, starch, and plant cell wall materials have been directly dissolved in the nondegrading solvent dimethylacetamide-lithium chloride (DMAC-LiCl) without prior isolation or extraction. Size-exclusion chromatography with viscometry and refractive index detectors was used with DMAC-LiCl as the mobile phase. The universal calibration concept was applied to obtain molecular weight distributions (MWDs). Applications include cotton fiber, corn and wheat starchflours,and avocado cell walls. Relationships were determined between the respective MWDs and cotton fiber variety, inheritance, textile processing, and strength; starch extrusion conditions; and avocado ripening stage.

-NATURAL

POLYMERS S U C H AS POLYSACCHARIDES, w h i c h usually have

h i g h molecular weight ( M W ) components, are difficult to characterize. A p p r o p r i a t e analytical techniques are generally dependent on getting the p o l y m e r into solution. Isolation and extraction often alter the p o l y mer composition (J). A v a i l a b l e solvents have serious limitations, most often because of degradation. M W standards are not generally available. T h e solvent d i m e t h y l a c e t a m i d e - l i t h i u m c h l o r i d e ( D M A C - L i C l ) offers the capacity for a w i d e a range of applications for dissolution of cellulose, starch, c h i t i n , and other polysaccharides w i t h little or no degradation f Deceased.

This chapter not subject to U . S . copyright. Published 1995 American Chemical Society

Provder et al.; Chromatographic Characterization of Polymers Advances in Chemistry; American Chemical Society: Washington, DC, 1995.

Downloaded by UNIV OF ALABAMA on November 21, 2016 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/ba-1995-0247.ch011

142

CHROMATOGRAPHIC CHARACTERIZATION OF POLYMERS

(2-4). C o t t o n fibers are single cells composed primarily (~96%) of the polymer cellulose. I n our laboratory (5), cotton fibers were dissolved directly i n the solvent D M A C - L i C l . This procedure solubilizes fiber cell wall components directly without prior extraction or derivatization, processes that could lead to degradation of high M W components. M W determinations have been carried out b y a size-exclusion chromatog raphy (SEC) system using commercial columns and instrumentation with D M A C - L i C l as the mobile phase. Incorporation of viscometry and r e fractive index (RI) detectors (6) allowed application of the universal calibration concept (7) to obtain M W distributions ( M W D s ) based on well-characterized narrow-distribution polystyrene standards (5). T h e universal calibration concept used b y incorporation of dual detectors bypasses the need for cellulose standards. There are no cellulose stan dards available. Polystyrene standards for a wide range of M W s dissolved readily i n D M A C - 0 . 5 % L i C l w i t h no activation necessary. W e extended the methodology developed for molecular character ization of cotton fiber to analysis of other polysaccharides. In this report, we present the results obtained from M W D s determined b y S E C for various complex carbohydrate samples dissolved i n D M A C - L i C l . A p plications include cotton fiber, corn and wheat starch flours, and avocado cell walls. Relationships are evaluated between the respective M W D s and cotton fiber development, variety, inheritance, textile processing, and strength; starch extrusion conditions; and stage of ripening i n avocado.

Experimental

Details

Safety Considerations. Ν,Ν-dimethylacetamide is an exceptional con tact hazard that may be harmful i f inhaled or absorbed through the skin and may be fatal to embryonic life in pregnant females (Baker Chemical C o m pany, Ν,Ν-dimethylacetamide, Material Safety Data Sheet, 1985, D 5 7 8 4 01; pp 1-4). Sample Preparation. Samples were dissolved as previously described (5). Ground material was added to D M A C (Burdick & Jackson, Muskegon, IL) in a Reacti-Vial (Pierce, Rockford, IL) in a heating block. Activation was achieved by elevating the temperature to 150 °C and maintained at that temperature for 1-2 h. The temperature was lowered to 100 °C followed by addition of dried L i C l (—8% wt/vol). Samples were held at 50 °C until dissolved (18-48 h) and subsequently were diluted and filtered. Final con centration of samples was 0.9-1.2 mg/mL in D M A C with 0.5% L i C l . A t least two dissolutions per sample were made for subsequent S E C analysis. Chromatography. Filtered polysaccharide solutions were analyzed using an S E C system consisting of an automatic sampler (Waters WISP, Waters, Milford, M A ) with a high-performance liquid chromatography pump (Waters model 590), pulse dampener (Viscotek, Houston, T X ) , viscometer


11.

Characterization by SEC with RI ir Viscometry

TiMPA

143

detector (Viscotek model 100), and RI detector (Waters model 410). The detectors were connected in series. The mobile phase was D M A C - 0 . 5 % L i C l pumped at a flow rate of 1.0 m L / m i n . Columns were Ultrastyragel 10 , 10 , ΙΟ , 10 (Waters) preceded by a guard column (Phenogel, linear, Phenomenex, Torrance, C A ) . A column heater (Waters Column Temperature System) regulated the temperature of the columns at 80 °C. Injection volume was 400 μL with a run time of 65 min. The software package Unical based on ASYST (Unical, Version 3.02, Viscotek) was used for data acquisition and analysis. Calibration was with polystyrene standards ranging in M W from 6.2 Χ 10 to 2.9 Χ 10 (Toyo Soda Manufacturing, Tokyo, Japan) dissolved and run in D M A C - 0 . 5 % L i C l . The universal calibration curve was a logarithmic function of the product of the intrinsic viscosity times M W versus retention volume with a third-order fit shown in Figure 1. 3

4

5

6


3

Results and

6

Discussion

D i s s o l u t i o n o f C o t t o n C e l l u l o s e . Attempts to identify the true M W of native cellulose always lead to difficulties, especially i n isolating unchanged celluloses from natural plant products and i n determining the M W of high M W celluloses by reliable physical methods (J). T h e Updegraff procedure is a frequently used method for measuring the

7.00

2.00

-I 25.0

1—

1

1 28.0

1

1

1 31.0

Ret. Vol

1

1

1 34.0

1

1

»— 37.0

1

1

1 40.0

(mL)

Figure 1. Universal calibration plot of polystyrene standards dissolved in DMAC-LiCl. (Log M) ΧIV is hydrodynamic volume, where M is molecular weight and IV is intrinsic viscosity.



144


cellulose content of plant material (8). Noncellulosic material is removed by treatment w i t h acetic a c i d - n i t r i c acid reagent at elevated temperature. T h e remaining cellulose is h y d r o l y z e d to glucose by treatment w i t h concentrated sulfuric acid. Glucose content of the sample is then determined colorimetrically. Often, only the first step of the procedure is used to remove the noncellulosic material. W e investigated the effects of the a c e t i c - n i t r i c reagent on the cellulosic composition. T h e M W D of cotton fiber exposed to a c e t i c - n i t r i c reagent was compared w i t h the native fiber sample. As indicated i n Table I, the effect of acid treatment was a shift of the M W D to a lower M W range corresponding to a 9 3 % reduction i n M W (9). After a c e t i c - n i t r i c reagent treatment, there is no evidence of low M W cellulose corresponding to polymers found i n the primary wall. This suggests that cellulose found i n the primary c e l l wall of cotton fiber is very susceptible to hydrolysis by the Updegraff reagent. Separate treatments of native cotton w i t h acid produced similar average values for M W , but the differences i n the broadness of the distribution of cellulosic chains indicates that the extent of polymer degradation is not exactly reproducible. Thus, direct solubilization of cotton fiber cell wall components i n D M A C - L i C l without prior extraction or derivatization avoids degradation of the polymer chains and is the preferred method for M W D determinations of cotton fiber cellulose. A p p l i c a t i o n s . Monitoring Cotton Fiber Development. Cotton fiber develops according to specific stages w i t h formation of a primary wall followed b y deposition of a secondary layer containing most of the cellulose (JO, J J). T h e biochemical composition of the fiber c e l l walls is changing throughout development; monitoring those changes has l i m ited progress i n this research area (12, 13). Cotton fiber begins development on the day flowering (anthesis); thus, the age of the fiber is usually designated b y days after flowering or days postanthesis ( D P A ) . T h e primary wall stage designates the rapid elongation of the outer c e l l wall occurring up to ^ 2 1 D P A , whereas the secondary wall development subsequently occurs with the major cellulose deposition. M a t u r e fiber T a b l e I.

T h e Effects o f A c i d T r e a t m e n t o n the M W o f N a t i v e C o t t o n F i b e r

Sample

h

D P /

Polydispersity

13,800

4900

2.8

1050

170

6.2

W

Native cotton fiber (Gossypium hirsutumh.) Texas Marker-I After treatment with acetic-nitric reagent a

DP °

DP is weight average of polymerization. DP is number average of polymerization. W N


11.

TiMPA

145

Characterization by SEC with RI b- Viscometry

is generally harvested from plants at —60 D P A . In our studies using S E C w i t h dual detectors (14), we observed that cell wall polymers from fibers at primary cell wall stages had lower M W s than the cellulose from fibers at the secondary wall stages, as shown i n F i g u r e 2. H i g h M W cellulose characteristic of mature cotton was detected as early as 8 D P A . H i g h M W material decreased during the p e r i o d of 1 0 - 1 8 D P A w i t h concomitant increase in lower M W wall components, possibly indicating hydrolysis during the later stages of elongation.


Monitoring

Cotton

Fiber

Inheritance

of

Molecular

Properties.

The

relationship between the M W D s , inheritance, and strength of cotton fiber in different genotypes was investigated (15, 16). F i b e r samples from a new higher (—10%) strength variety for the Mississippi D e l t a (Mississippi D e l t a 5 1 , M D 5 1 ne) were assayed and compared w i t h a popular commercial variety and recurrent parent (Deltapine 90). T h e F l cross and two selected backcrossed progenies were evaluated from three replications g r o w n i n the field. The fiber samples had very similar genetic backgrounds and physical properties except for strength. S E C w i t h viscometry and R I was used to determine the M W D of cotton fiber samples dissolved directly in D M A C - L i C l . As shown i n Figure 3, samples of the higher strength variety had a greater proportion of higher M W material than the commercial variety. T h e most significant difference was apparent i n the M W range of 1,000,000-10,000,000. 6.00

5.00

4.00

Ο X

3.00

2.00

1 .00

.000 2.00

3.00

4.00

5.00

6.00

7.00

β.00

LOG M

Figure 2. MWDs of samples of cotton fiber at different stages of devel opment: 10 DPA or primary wall stage versus 60 DPA or mature fiber.


146

CHROMATOGRAPHIC CHARACTERIZATION OF POLYMERS 14.0

τ

Molecular Weight Distribution

11.0

+

8.00

+

5.00

+

2.00

+

ο


χ

-1.00 2.00

3.00

4.00

5.00

6.00

7.00

Log M

Figure 3. Comparison of different varieties of cotton by MWDs of samples of mature cottonfiber.Mississippi Delta 51 (MD 51 ne) has ~ 10% higher fiber strength than Deltapine 90.

Effects of Textile Processing. Mercerization has been used to impart desirable qualities to cotton fiber for many years. Changes i n crystalline structure of cotton cellulose are one of the major results. I n our labo ratory (17), alterations i n the molecular composition of cotton fiber caused by mercerization was measured by S E C . Samples from Deltapine, Acala, and P i m a varieties were evaluated as scoured or as mercerized fiber. S E C determinations of the M W D s of cotton fiber samples before and after caustic mercerization showed significant loss i n the higher M W fractions for each of the three varieties. Comparison of M W D s for the acala cotton sample and corresponding mercerized treatment are provided i n F i g u r e 4. M e r c e r i z a t i o n b y l i q u i d ammonia also affected the higher M W components, although changes in molecular composition were different from that observed w i t h caustic mercerization. Effects of Extrusion on Starch. Processing starch b y extrusion r e sults i n molecular fragmentation. T h e effects on the M W D s of flours from wheat and corn starch were determined (18-20). Starch flours var ied i n amylose, amylopectin, and protein content. Samples were sub jected to twin-screw extrusion w i t h varying moisture content, screw speed, die temperature, mass flow rate, and protein content. Starch flours were directly dissolved i n the solvent D M A C - L i C l without prior iso-


11.

TlMPA

16.0

Characterization by SEC with RI 10 (%)

MW 10 -10 (%)

18.8 7.4 6.0 7.3 9.4

26 13 8 17 14

26 27 29 26 26

7

6


7

(


148


4

·

0 0

5.00

6.00

7.00

θ . 0 0

9.00

Log M

Figure 5. Cumulative MWDs of the high protein wheatflourcontrol and an extruded sample (G39).

sessments of branching by monitoring the intrinsic viscosity have been carried out. F o r example, shifts i n M a r k - H o u w i n k plots are evident for branched samples versus linear types (Figure 6). Monitoring Fruit Ripening and Cell-Wall Turnover. Avocados are a popular fruit for market. T h e desirability depends on the softening of the flesh of the fruit. O u r research identified key components i n the composition of the avocado c e l l wall and the changes that occur during stages of ripening (21). T h e cellulose molecular structure and crystalline association of cell walls during fruit ripening was monitored and related to the levels of the enzyme cellulase. Cellulase is an enzyme that spe cifically degrades cellulose. S E C techniques used to study cotton fiber molecular structure were used to evaluate avocado cellulose, and X - r a y diffraction and electron microscopy were used to look at the cellulose fibers. S E C of total cell wall polysaccharides (including cellulose) r e vealed a slight increase i n the fraction of the largest polymers during ripening, whereas the crystallinity index increased. Based on these r e sults, we propose that the cellulase prefers to attach the noncrystalline portions of the cellulose i n the wall. This mode of action affects the firmness of the avocado fruit during the ripening process.


11.

TlMPA

8.00

Characterization by SEC with RI ir Viscometry

149

χ M a r k - H o u w i n k Plot


4.00

f

-16.0 2.00

High A m y l o s e C o r n

3.00

4.00

5.00

6.00

7.00

8.00

9.00

Log M

Figure 6. Mark-Houwink plots of intrinsic viscosities of starch standards (amylose and amylopectin) compared with wheat and cornflours.

Summary Dissolution of whole plant cell walls or commercially important complex carbohydrates directly into D M A C - L i C l was advantageous. M o l e c u l a r characterization by S E C w i t h viscometry and R I detectors using the universal calibration allowed evaluation not readily attainable previously (22). Determination of the M W D s of c e l l wall polymers at critical stages of development provides a tool for understanding biological regulation of the growth processes i n cotton fiber and avocado. I n addition, m o n itoring effects of commercial processing of natural polymers assists i n minimizing losses and i m p r o v e d end-use products.

Acknowledgments I thank the following people for their cooperation: D . J . H u b e r , W . R . M e r e d i t h , E . M . O ' D o n o g h u e , M . P o l i t z , H . H . Ramey, B . A . Triplett, B. P . Wasserman, A . Striegel, and S. H . Zeronian.

References 1. Franz, G.; Blaschek, W . In Methods in Plant Biochemistry; Dey, P. M . ; Harbrone, J., Eds.; Academic: Orlando, FL, 1990; Vol. 2, pp 291-322.


150

CHROMATOGRAPHIC

CHARACTERIZATION OF POLYMERS

2. Turbak, A . B . In Wood and Agricultural Residues; Soltes, E.J., Ed.; Academic: O r l a n d o , FL, 1 9 8 3 ;

pp

87-99.

3. M c C o r m i c k , C . L.; Callais, P. Α.; H u t c h i n s o n , Β. H. Macromolecules 1985, 18,

2394-2401.

4. Dawsey, T. R.; M c C o r m i c k , C . L. Rev. Macromol. Chem. Phys. 1990,

C30,

403-440.

5. T i m p a , J . D. J. Agri. Food Chem. 1991, 3 9 , 6.

H a n e y , M. A . A m . Lab.

1985,

17,

270-275.

116-126.

7. G r u b i s i c , A.; R e m p p , P . ; Benoit, H. A . Polym. Lett. 1967, 5, 753-759. 8. Updegraff, D . M. Anal. Biochem. 1969, 32, 420-424. 9. T i m p a , J . D.; T r i p l e t t , B . A . Plant Physiol. (Life Sci. Adv.) 1992, 11, 253-


256.

10.

M a r x - F i g i n i , M. In Cellulose and Other Natural Polymer Systems: Biogenesis, Structure, and Degradation; B r o w n , R. M., Ed.; P l e n u m : N e w Y o r k , 1 9 8 2 ;

11. 12. 13. 14. 15.

18. 19.

Basra, A. S.; M a l i k , C . P. Int. Rev. Cytol. 1984, 89, 65-113. M e i n e r t , M. C . ; D e l m e r , D . P . Plant Physiol. 1977, 5 9 , 1 0 8 8 - 1 0 9 4 . D e l m e r , D . P. Annu. Rev. Plant Physiol. 1987, 38, 259-292. T i m p a , J. D.; T r i p l e t t , B . A . Planta 1993, 189, 1 0 1 - 1 0 8 . T i m p a , J. D. In Cotton Fiber Cellulose: Structure, Function and Utilization; National C o t t o n C o u n c i l : M e m p h i s , TN, 1 9 9 3 ; pp 376-382. T i m p a , J . D.; M e r e d i t h , W . R. Proceedings of the Beltwide Cotton Production Research Conferences; National C o t t o n C o u n c i l : M e m p h i s , TN, 1 9 9 3 ; V o l . 3, p 1 5 5 6 . T i m p a , J. D.; Zeronian, S. H. Proceedings of the Beltwide Cotton Production Research Conferences; National C o t t o n C o u n c i l : M e m p h i s , TN, 1 9 9 3 ; V o l . 3, p 1 4 9 3 . Wasserman, B . P . ; T i m p a , J. D. Starch/Starke 1991, 43, 389-392. P o l i t z , M.; T i m p a , J . D.; Wasserman, B . R. Cereal. Chem. 1994, 71, 532-

20.

P o l i t z , M.; T i m p a , J. D.; W h i t e , A . R.; Wasserman, B . R. Carbohydr. Polym.

21. 22.

O ' D o n o g h u e , E . M.; H u b e r , D . J.; T i m p a , J. D. Planta 1994, 194, T i m p a , J. D. Trends Polym. Sci. 1993, 1, 105-110.

pp

16.

17.

243-271.

536. 1994,

24,

91-99.

573-584.

RECEIVED for review January 6, 1 9 9 4 . ACCEPTED revised manuscript A p r i l 2 6 , 1994.


Characterization by Size-Exclusion Chromatography with Refractive

Recommend Documents