N‐Terminal Amino Acid Sequence Analysis of Endosperm Proteins in Japanese Hexaploid Wheat

Semidry electroblotting is convenient and allows a rapid and efficient protein transfer from two‐dimensional polyacrylamide gel electrophoresis (2D‐PAGE) gels onto sequencer stable supports for protein microsequence analysis in a gas‐phase sequencer. Using this technique, I determined the amino acid sequences of the endosperm proteins in Japanese hexaploid commercial wheats (Triticum aestivum). Based on sequence determination of the Japanese hexaploid wheats, the endosperm protein could be easily characterized. Wheat endosperm protein, extracted in the presence of 2‐mercaptoethanol and SDS, fractionated into many protein polypeptides using 2D‐PAGE under dissociating conditions. These components were grouped into HMW glutenin subunits, α‐, β‐ or γ‐gliadins, and novel protein polypeptides by using the N‐terminal amino acid sequences. The novel endosperm protein polypeptides were detected, and two new types of N‐terminal amino acid sequences have been found for protein poly‐peptides. These polypeptides have much faster electrophoresis mobility during 2D‐PAGE and are therefore probably a much smaller size than any other peptides of endosperm protein groups found in hexaploid wheat. Ten protein polypeptides have been purified from cultivars of Japanese wheat. Some differences in the contents of amino acids for four protein polypeptide spots were apparent in Japanese wheat.     

Molecular Weight Distribution of Proteins in Hard Red Spring Wheat: Relationship to Quality Parameters and Intrasample Uniformity

Molecular weight distribution (MWD) of proteins extracted from hard red spring wheat was analyzed by size‐exclusion HPLC to investigate associations with wheat and breadmaking quality characteristics. Certain protein fractions were related to associations between wheat and breadmaking parameters, specifically when effect of quantitative variation of protein on those parameters was statistically eliminated by partial correlation analysis. SDS‐unextractable high molecular weight polymeric proteins had positive partial correlations with percent vitreous kernel content and breadmaking parameters, including mix time and bread loaf volume. SDS‐extractable protein fractions that were eluted before the primary gliadin peak had positive partial correlations with kernel hardness and water absorption parameters. The proportion of main gliadin fractions in total protein had a negative partial correlation with bread loaf volume and positive correlations with kernel hardness and water absorption parameters. Intrasample uniformity in protein MWD and kernel characteristics was estimated from three kernel subsamples that were separated according to single kernel protein content within individual wheat samples by a single‐kernel near‐infrared sorter. Wheat subsamples were significantly different in protein MWD. Intrasample uniformity in protein MWD did not differ greatly among wheat samples.     

Compressive Strength of Wheat Endosperm: Comparison of Endosperm Bricks to the Single Kernel Characterization System

The three major classes of endosperm texture (grain hardness) of soft and hard common, and durum wheat represent and define one of the leading determinants of the milling and end‐use quality of wheat. Although these three genetic classes are directly related to the Hardness locus and puroindoline gene function, much less is known about the kernel‐to‐kernel variation within pure varietal grain lots. Measurement of this variation is of considerable interest. The objective of this research was to compare kernel texture as determined by compression failure testing using endosperm bricks with results of whole‐kernel hardness obtained with the Single Kernel Characterization System 4100 hardness index (SKCS HI). In general terms, the variation obtained with the SKCS HI was of similar magnitude to that obtained using failure strain and failure energy of endosperm brick compression. Objective comparisons included frequency distribution plots, normalized frequency distribution plots, ANOVA model R², and coefficients of variation. Results indicated that compression testing and SKCS HI similarly captured the main features of texture classes but also reflected notable differences in texture properties among and within soft, hard, and durum classes. Neither brick compression testing nor the SKCS HI may be reasonably expected to correctly classify all individual kernels as to genetic texture class. However, modest improvements in correct classification rate or, more importantly, better classification related to end‐use quality may still be achievable.     

Association of Size‐Exclusion HPLC of Endosperm Proteins with Dough Mixing and Breadmaking Characteristics in a Recombinant Inbred Population of Hard Red Spring Wheat

Variation of polymeric proteins affects wheat end‐use quality. This research investigated associations of polymeric proteins with dough mixing strength and breadmaking characteristics in a near‐homogenous population of 139 recombinant inbred lines (RILs) derived from a cross between two hard red spring wheat breeding lines. Flours from the RILs grown at three locations were analyzed for molecular weight (MW) distribution of SDS‐extractable and unextractable proteins using size‐exclusion HPLC protocol. Correlations were calculated between mixing and breadmaking properties and HPLC absorbance data obtained a 0.01‐min retention time interval to identify protein fractions that had a significant effect on the quality traits. Very high MW polymeric proteins in the unextractable fraction had more distinct and positive associations with dough mixing strength and bread loaf volume than did other polymeric protein fractions, whereas extractable polymeric had negative influence. Consequently, the ratio of unextractable very high MW polymeric proteins to extractable polymeric proteins had greater correlations with dough mixing parameters than other HPLC absorbance area data. Covariate‐effect biplots also visually validated positive effects of unextractable very high MW polymeric proteins and negative effects of extractable polymeric proteins on mixing properties and loaf volume across three growing locations.     

Differences in the Aleurone Layer Fate Between Hard and Soft Common Wheats at Grain Milling

In the milling process, efficient separation between the starchy endosperm and the other grain tissues is a key parameter estimated by ash measurement. Because this separation occurs near the aleurone layer interface, better understanding of this tissue fractionation is critical for a better analysis of the wheat milling behavior. Samples from hard and soft common wheat cultivars that had the same protein content were processed on a pilot mill, and whole grain meals or flour streams were analyzed for ash content. The para‐coumaric acid (p‐CA) and phytic acid flour contents were compared with ash measurement and used as markers of the aleurone cell walls or aleurone cell content, respectively. A greater amount of phytic acid in hard wheat flour compared with soft wheat flour was found and reveals a distinct milling behavior between those wheat classes, mainly at the breaking step. Therefore simple ash content measurement is not sufficient to analyze flour purity. At the reduction stage, quantity of phytic acid increases with the other markers and may result from the overall mechanical resistance of the aleurone tissue. As a consequence, wheat hardness not only determines grain milling behavior but also affects flour composition.     

Rapid Quality Assessment of Wheat Cultivars Registered in Poland Using the 2‐g Mixograph and Multivariate Statistical Analysis

Baking and 2‐g mixograph analyses were performed for 55 cultivars (19 spring and 36 winter wheat) from various quality classes from the 2002 harvest in Poland. An instrumented 2‐g direct‐drive mixograph was used to study the mixing characteristics of the wheat cultivars. A number of parameters were extracted automatically from each mixograph trace and correlated with baking volume and flour quality parameters (protein content and high molecular weight glutenin subunit [HMW‐GS] composition by SDS‐PAGE) using multiple linear regression statistical analysis. Principal component analysis of the mixograph data discriminated between four flour quality classes, and predictions of baking volume were obtained using several selected mixograph parameters, chosen using a best subsets regression routine, giving R² values of 0.862–0.866. In particular, three new spring wheat strains (CHD 502a‐c) recently registered in Poland were highly discriminated and predicted to give high baking volume on the basis of two mixograph parameters: peak bandwidth and 10‐min bandwidth.     

Size‐Exclusion HPLC of Protein Using a Narrow‐Bore Column for Evaluation of Breadmaking Quality of Hard Spring Wheat Flours

The objective of this study was to investigate whether a narrow‐bore column (NBC) (300 × 4.5 mm, i.d.) improved analyses of unreduced proteins in flour by size‐exclusion HPLC (SE‐HPLC) and subsequent evaluation of breadmaking quality of hard spring wheat flours. Total protein extracts and SDS buffer extractable and unextractable proteins were analyzed by SE‐HPLC. NBC separated proteins in 10 min at a flow rate of 0.5 mL/min with similar resolution to a regular column (300 × 7.8 mm, i.d.) which took 30 min. SE‐HPLC absorbance area (AA) data obtained from an NBC showed comparable or superior repeatability and correlations with flour breadmaking characteristics when compared with those of a regular column. AA values of total protein that were calculated by adding AA values of SDS extractable and unextractable proteins showed greater repeatability and correlations with quality characteristics than those of actual total protein extracts. The improvements including employment of an NBC in SE‐HPLC provide enhancement of rapid quality evaluation and decreased consumption of hazardous organic solvents.     

Improved HPLC method for determination of phytosiderophores in root washings and tissue extracts

Phytosiderophores (PS) of the mugineic acid family can be separated effectively by HPLC on resin‐based anion exchange columns. Using gradient elution with aqueous NaOH, separation of 2'‐deoxymugineic acid (DMA), mugineic acid (MA), 3‐hydroxymugineic acid (HMA), 3‐epi‐hyydroxymugineic acid (epi‐HMA), and nicotianamine was obtained within 16 min with a complete cycle time of 30 min. Fluorimetric detection was performed after post‐column derivatization using sodium hypochlorite (NaOCl) and orthophtaldialdehyde. External standardization revealed a linear range between 0.1–2.5 nmol of each compound applied to the column, with a detection limit of approximately 0.05 nmol. Only minimal sample pre‐treatment by addition of sodium hydroxide (NaOH) was required prior to HPLC‐injection of natural occurring samples such as root exudates, collected from the whole root system or from single apical root zones, xylem sap, and hot water extracts of root material, obtained from iron (Fe)‐deficient maize and barley plants. The method is discussed in comparison with cation exchange HPLC which is conventionally employed for the separation of phytosiderophores.     

Hydration Dynamics of Durum Wheat Endosperm as Studied by Magnetic Resonance Imaging and Soaking Experiments

The precise knowledge of the kinetics of water transport in durum wheat endosperm is a prerequisite for the optimization of wheat processing techniques like pasta dough mixing on a fundamental basis. Pieces of endosperm were cylindrically cut, prepared from durum wheat kernels, and used to study the water uptake by applying a gravimetric method and magnetic resonance imaging (MRI). The total water uptake of endosperm cylinders at different soaking times was determined by gravimetric soaking experiments and revealed a swelling limit of ≈40 g/100 g wb after 60 min. With these results it was possible to estimate an apparent diffusion coefficient of water in durum endosperm by using numerical simulation based on a diffusion model (D_25°C ~ 0.76 × 10^–10 m²/sec). MRI was used to quantify the water distribution in the endosperm cylinders over time at excess and limited water conditions. The calibration of MRI for the quantification of local and time‐dependent water contents was successful by correlating the spin‐spin relaxation time (T₂) with the water content of calibration samples at intermediate moisture levels (19–45 g/100 g wb). Water content maps were generated and showed the kinetics of water distribution inside the endosperm cylinders up to equilibrium conditions. The water uptake of the endosperm cylinders over time, as measured by MRI, fitted well to the water uptake as determined gravimetrically in soaking tests, which validated the applied MRI calibration and measurement procedures. The results allow the quantitative prediction of water transport properties of durum wheat endosperm during moistening procedures.     

Mechanical Properties of Wheat Seed Coats

The crumbliness of starchy endosperm and the resistance of bran are key characteristics that enhance milling behavior of wheat and are dependent on the genetic origin and moisture content of the grain. A method was developed to measure the mechanical properties of bran samples based on the measurement of tensile stress and strain. Tests conducted with this highly reproducible and sensitive method documented cultivar and moisture‐content effects (6.3, 13.8, and 18%, wb) on rheological behavior of wheat seed coats. A moisture‐dependent reduction in stress to fracture (‐15 to ‐30%) and in Young's modulus (‐45 to ‐55%) was quantified. An increase in deformation to fracture of seed coats was also correlated with bran size differences after milling. The energy required to fracture a sample (from 0.4 to 1.3 J/mm³) was considered the most valid of all presented parameters for assessing the milling behavior of wheat seed coats and the size of bran fractions.     

Procedure for Obtaining Stable Protein Extracts of Cereal Flour and Whole Meal for Size‐Exclusion HPLC Analysis

Experiments were conducted to determine the extent of instability of size‐exclusion HPLC extracts prepared from flour, semolina, or whole meal. Procedures to obtain stable extracts were investigated. Whole meal extracts of durum wheat and triticale were the most unstable samples, whereas bread wheat showed smaller changes. Samples prepared from crushed maturing grains, especially during early stages, were greatly influenced by the instability process. By using protease inhibitors, evidence was obtained that endogenous proteases were the source of the instability. Reproducible peak 1 (polymeric protein) results were obtained for a period of at least 72 hr after extract preparation if protein extracts were heated for 2 min at 80°C in a water bath immediately after filtration into the sample vials and before SE‐HPLC analysis. This treatment is a viable solution to avoid sample instability in whole meal and developing grain extracts, particularly when large sample sets are prepared for automatic injection into the HPLC.     

Rapid High‐Performance Liquid Chromatography Determination of Tocopherols and Tocotrienols in Cereals

A rapid normal‐phase high‐performance liquid chromatography (NP‐HPLC) method has been developed for rapid determination of the total content of tocopherols (T) and tocotrienols (T3) in cereals. The new method uses a one‐step extraction followed by NP‐HPLC coupled with a fluorescence detector. The new method provides a baseline separation of the critical peaks of beta‐tocotrienol (β‐T3) and gamma‐tocopherol (γ‐T) within a short time of analysis. The extraction step requires no saponification step or addition of antioxidant. Addition of butylated hydroxytoluene (BHT) had no effect on the stability of vitamin E during sampling. The recovery was 96–100%. The method is demonstrated by successful quantification of vitamin E in barley, wheat, and spelt grains, as well as wheat germ and flours from wheat, spelt, and rye.     

Quantitation of LMW‐GS to HMW‐GS Ratio in Wheat Flours

A comparison was made of methods for measuring the LMW/HMW glutenin subunit (GS) ratio for glutenin. A set of near‐isogenic wheat lines with the number of HMW‐GS varying from 0 to 5 was utilized to provide a wide range of LMW/HMW‐GS. Glutenin preparations were obtained from ground whole meal after solubilization of monomeric proteins by dimethyl sulfoxide (DMSO) or 50% propanol or by fraction collection from a preparative SE‐HPLC column. Analyses were made on the reduced glutenin from each of the three preparations by RP‐HPLC, SE‐HPLC, and SDS‐PAGE. Both solvents, DMSO and 50% propanol, extracted appreciable amounts of polymeric protein, thus casting some doubts on the accuracy of the determinations. This problem was largely avoided when the polymeric fraction was collected from the eluate of a total glutenin extract run on a preparative SE‐HPLC column. Less glutenin was removed by the two solvents for lines with a greater number of HMW‐GS or with strength‐associated HMW‐GS 5+10 coded by the 1D chromosome. Collection of the polymeric protein in SE‐HPLC, followed by separation of the glutenin subunits in RP‐HPLC, was the best method for quantitating the LMW/HMW‐GS ratio. SE‐HPLC gave a clear separation of the two groups of subunits as well as HMW albumins. RP‐HPLC has the potential advantage of being able to quantitate individual subunits.     

Mechanical and Physicochemical Characterization of Vitreous and Mealy Durum Wheat Endosperm

The mechanical, physical, and biochemical characteristics of mealy and vitreous endosperm were investigated. Endosperm were obtained from four durum wheat cultivars grown under different nitrogen fertilization designs. The textural properties and the density of the endosperm were measured on hand‐shaped parallelepiped endosperm samples. Endosperm protein content and composition and also gliadin composition were investigated by HPLC. Mechanical tests showed that mealy and vitreous endosperm differed in hardness and vitreousness. Vitreousness increased with nitrogen fertilization supply whereas there was no variation among the different cultivars. Hardness seemed to be linked to genotype and insensitive to nitrogen supply. From this result, we concluded that hardness and vitreousness are not related. Endosperm protein content and gliadin‐to‐glutenin ratio were related to nitrogen supply and increased especially when nitrogen supply was applied at flowering. At the same time, endosperm vitreousness increased. Further biochemical analyses were performed on 270 kernels, mealy or vitreous, hand‐picked from 148 different crops. Results showed that protein content of vitreous endosperm exceeded 9.7% in >90% of the cases. The glia/glu ratio was a less accurate predictor of kernel vitreousness, indicating that, by itself, it cannot account for the change in kernel vitreousness. Endosperm vitreous texture would rise above a threshold content of 9.7% protein within the endosperm.     

Simplified Dilute Acetic Acid Based Extraction Procedure for Fractionation and Analysis of Wheat Flour Protein by Size Exclusion HPLC and Flow Field‐Flow Fractionation

A simple, highly efficient and reproducible two‐step extraction procedure using dilute acetic acid without (AN) and then with sonication (AS) has been developed for the fractionation of wheat flour protein. Approximately 97% of total protein was extracted from a Canadian hard red spring wheat flour; an additional 1.2% protein could be recovered by further extraction with 1% DDT and 50% 1‐propanol (AR). Size‐exclusion HPLC (SE‐HPLC) and flow field‐flow fractionation (flow FFF) showed that the AN extract, which accounted for most of the total extractable protein (AN + AS + AR), consisted primarily of monomeric protein. The AS extract was composed primarily of polymeric proteins. Flow FFF showed that AN polymeric protein, including that eluting at the SE‐HPLC void volume, showed smaller Stokes diameters than AS polymeric protein. Flow FFF profiles of AS SE‐HPLC subfractions showed that the void volume subfraction contained monomeric and small polymeric protein in addition to large polymeric protein, indicating formation of larger complexes through interaction between some or all of the components. AN and AS extracts, as well as SE‐HPLC and flow FFF fractions thereof, showed a fairly wide range of values among 12 Canadian hard red and white spring wheat cultivars. The proportion of total protein in the AS extract and in the larger sized polymeric protein fractions from SE‐HPLC and flow FFF were highly positively correlated to farinograph mixing time.     

Grain Quality Characteristics and Milling Performance of Full and Partial Waxy Durum Lines

Mutation of the gene coding for the granule bound starch synthase (waxy protein) leads to reduced amylose content in cereal endosperm. Durum wheat (Triticum turgidum L. var. durum) has one waxy locus in each of its two genomes. Full waxy durum wheat is produced when both genomes carry the waxy null alleles. When only one locus is mutated, partial waxy durum wheat is obtained. Partial and full waxy near‐isogenic lines of durum wheat developed by a breeding program were analyzed as to their quality characteristics. Amylose was largely eliminated in full waxy lines; however, no reduction in amylose content was detected in partial waxy lines. The waxy mutation did not affect grain yield, kernel size, or kernel hardness. Full waxy durum lines had higher kernel ash content, α‐amylase activity, and a unique nonvitreous kernel appearance. Protein quality, as evaluated by SDS microsedimentation value, gluten index, and wet gluten was slightly lower in the full waxy lines than in the other genotypes. However, comparisons with current cultivars indicated that protein quality of all derived lines remained in the range of strong gluten cultivars. Semolina yield was lowered by the waxy mutations due to lower friability that resulted in less complete separation of the endosperm from the bran. Waxy semolina was more sensitive to mechanical damage during milling, but modified tempering and milling conditions may limit the damage. Overall, quality characteristics of waxy durum grain were satisfactory and suitable for application testing.     

Image Analysis Approach to Understand the Differences in Flour Particle Surface and Shape Characteristics

The separation efficiency of wheat flour particles based on size, with minimum bran contamination, is important for a flour mill. Separation of flour during fractionation depends on the surface characteristics and shape of flour particles. Wheat flour particle characteristics such as surface lipid content, roughness, and morphology with respect to particle size were studied to better understand the differences between hard and soft wheat flours. Fractal analysis using image analysis was used to ascertain surface roughness. That was in turn verified by atomic force microscopy measurements. Soft wheat flours (soft red winter and soft white) had a higher degree of surface roughness than the hard wheat flours (hard red spring, hard red winter, and hard white). The fractal dimension values ranged from 2.67 to 2.78 and from 2.28 to 2.55 for soft and hard wheat flours, respectively. The surface lipid content increased with particle size in hard wheat but decreased in soft wheat flours. The surface lipid levels ranged from 1.02 to 1.18 and from 2.55 to 2.58% (% of total area) for 45 μm particles in hard wheat flours (hard red spring, hard red winter, and hard white) and soft wheat flours (soft red winter and soft white), respectively. For the 90 μm particles the lipid levels ranged from 1.54 to 1.62 and from 1.70 to 1.83% (% of total area) for flour particles in hard wheat flours (hard red spring, hard red winter, and hard white) and soft wheat flours (soft red winter and soft white), respectively. Surface lipid content and roughness values showed that soft wheat flours will be more cohesive than hard wheat flours. The morphology values revealed the irregularity in flour particles, irrespective of wheat class and particle size, owing to nonuniform fragmentation of protein and starch matrix of the wheat endosperm.     

Collaborative Analysis of Wheat Endosperm Compressive Material Properties

The objective measurement of cereal endosperm texture, for wheat (Triticum spp. L.) in particular, is relevant to the milling, processing, and utilization of grain. The objective of this study was to evaluate the interlaboratory results of compression failure testing of wheat endosperm specimens of defined geometry. Parallelepipeds (bricks) and cylinders were prepared from individual soft and hard near‐isogenic wheat kernels and compressed in two orientations (parallel and perpendicular to the long brush‐to‐germ axis). Compression curves were used to derive failure stress, failure strain, work density (area under the curve), and Young's modulus. In all five laboratories, the ability to delineate hard from soft wheat endosperm material properties was quite high. Four laboratories compressed endosperm bricks in the same orientation, on edge; texture class (soft vs. hard) was consistently the greatest source of variation in analysis of variance models (F‐values from 417 to 1401, Young's modulus and failure stress, respectively). Failure stress was found to be the best overall means of measuring the difference in what is known in the vernacular as wheat hardness. Across laboratories, the absolute measures of all four material properties ranged on the order of about two‐ to threefold from low to high, although within a laboratory, results were highly consistent. Laboratory by texture class interaction was deemed to be of minor importance. Brick size and moisture content within the ranges tested were not major sources of variation, and cylinders prepared from endosperm produced results similar to those obtained from bricks. The results suggested that wheat endosperm might express some level of anisotropic behavior, as specimens compressed in the kernel orientation parallel to the long axis failed at lower strain and stress values, with lower work density, when compared with kernel orientation perpendicular to the long axis. A key feature of interlaboratory variation was identified as being instrument rigidity, a subject of ongoing research. In conclusion, the preparation of endosperm specimens of defined size and shape, in combination with compression failure testing at low moisture content (<18%), is useful for objectively delineating the phenomenon known as hardness. The study presented here will advance our ability to objectively measure cereal grain texture and the material properties of endosperm.     

Evaluation of a Near‐Infrared Reflectance Spectrometer as a Granulation Sensor for First‐Break Ground Wheat: Studies with Hard Red Winter Wheats

A single wheat class or blended wheats from two wheat classes are usually milled in a flour mill. A near‐infrared (NIR) reflectance spectrometer, previously evaluated as granulation sensor for first‐break ground wheat from six wheat classes, was evaluated for a single wheat class, hard red winter (HRW) wheat, using offline methods. The HRW wheats represented seven cultivars ground by an experimental roller mill at five roll gap settings (0.38, 0.51, 0.63, 0.75, and 0.88 mm) which yielded 35 ground wheat samples each for the calibration and validation sets. Granulation models based on partial least squares regression were developed with cumulative mass of size fractions as a reference value. Combinations of four data pretreatments (log 1/R, baseline correction, unit area normalization, and derivatives) and subregions of the 400–1,700 nm wavelength range were evaluated. Models that used pathlength correction (unit area normalization) predicted well each of the four size fractions of first‐break ground wheat. The best model, unit area normalization and first derivative, predicted all the validation spectra with standard errors of performance of 3.80, 1.29, 0.43, and 0.68 for the >1041, >375, >240, and >136 μm size fractions, respectively. Ground HRW wheats have narrower particle size distribution and better sieving properties than ground wheat from six wheat classes. Thus, HRW wheat granulation models performed better than the previously reported models for six wheat classes.     

Impact of Cultivar and Environment on Size Characteristics of Wheat Proteins Using Asymmetrical Flow Field‐Flow Fractionation and Multi‐Angle Laser Light Scattering

The use of multi‐angle laser light scattering (MALLS) in conjunction with asymmetrical flow field‐flow fractionation (A‐FFFF) was investigated for the determination of the molecular weight distribution (MWD) of wheat proteins. The wheat flour proteins were dissolved by sonication in 0.1M sodium phosphate (pH 6.9) containing 2% SDS. The results presented make it evident that efficient separation and size characterization of monomeric (M < 10⁵ g/mol) and polymeric protein (10⁵ ≤ M < 10⁸ g/mol) wheat proteins can be achieved with A‐FFFF/MALLS/UV in a single run. Therefore, this method appears to be able to detect significant modifications of MWD of wheat protein, whatever the factor inducing these alterations (i.e., genetic or environmental) and whatever the nature of these alterations (i.e., monomeric‐to‐polymeric ratio or MWD of polymeric protein). In the present study, we have indeed demonstrated that the MWD of wheat proteins can be altered from one cultivar to another in three main ways: by changing the relative amounts of monomeric and polymeric proteins, by changing the MWD of polymeric protein, and then by changing both the monomeric‐to‐polymeric ratio and the MWD of polymeric protein.