Accepted Manuscript Determination of the Degree of Acetylation and the Distribution of Acetyl Groups in Chitosan by HPLC Analysis of Nitrous Acid Degraded and PMP Labeled Products Zhangrun Han, Yangyang Zeng, Hong Lu, Lijuan Zhang PII:

S0008-6215(15)00078-6

DOI:

10.1016/j.carres.2015.03.002

Reference:

CAR 6956

To appear in:

Carbohydrate Research

Received Date: 17 January 2015 Revised Date:

27 February 2015

Accepted Date: 3 March 2015

Please cite this article as: Han Z, Zeng Y, Lu H, Zhang L, Determination of the Degree of Acetylation and the Distribution of Acetyl Groups in Chitosan by HPLC Analysis of Nitrous Acid Degraded and PMP Labeled Products, Carbohydrate Research (2015), doi: 10.1016/j.carres.2015.03.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Determination of the Degree of Acetylation and the Distribution of Acetyl Groups in

Zhangrun Han‡; Yangyang Zeng‡; Hong Lu; and Lijuan Zhang*

RI PT

Chitosan by HPLC Analysis of Nitrous Acid Degraded and PMP Labeled Products

School of Medicine and Pharmacy, Ocean University of China, Qingdao, China

SC

*Corresponding author: School of Medicine and Pharmacy, Ocean University of China, 5 Yushan

[email protected]

AC C

EP

TE D

‡These authors contributed equally.

M AN U

Road, Qingdao, Shandong Province, China 266003; Tel.: +86 0532-82031615; E-mail:

1

ACCEPTED MANUSCRIPT

ABSTRACT

Chitin is one of the most abundant polysaccharides on earth. It consists of repeating β-1, 4 linked

RI PT

N-acetylated glucosamine (A) units. Chitosan is an N-deacetylated product of chitin. Chitosan and its derivatives have broad medical applications as drugs, nutraceuticals, or drug delivery agents. However, a reliable analytical method for quality control of medically used chitosans is still lacking.

SC

In current study, nitrous acid was used to cleave all glucosamine residues in chitosan into 2, 5anhydromannose (M) or M at the reducing end of di-, tri-, and oligosaccharides. PMP, i.e. 1-phenyl-

M AN U

3-methyl-5-pyrazolone, was used to label all the Ms. Online UV detection allowed quantification of all M-containing UV peaks whereas online MS analysis directly identified 11 different kinds of mono-, di-, tri-, and oligosaccharides that correlated each oligosaccharide with specific UV peak after HPLC separation. The DA (degree of acetylation) for chitosans was calculated based on the A/(A+M) value derived from the UV data. This newly developed method had several advantages for

TE D

quality control of chitosan: 1. the experimental procedures were extensively optimized; 2. the reliability of the method was confirmed by online LC-MS analysis; 3. the DA value was obtainable based on the UV data after HPLC analysis, which was comparableto that of 1H-NMR and

EP

conductometric titration analyses; 4. finally and most importantly, this method could be used to obtain the DA as well as chemical acetylation/deacetylation mechanisms for

chitosan by any

AC C

laboratory equipped with a HPLC and an online UV detector.

Keywords: degree of acetylation; chitosan; distribution of acetyl groups in chitosan; nitrous acid (HONO); 2, 5-anhydromannose (M); 1-phenyl-3-methyl-5-pyrazolone (PMP).

2

ACCEPTED MANUSCRIPT 1. Introduction

Chitin is the second most abundant biopolymer on Earth synthesized and degraded by

annually.

[1]

RI PT

marine invertebrate, insects, fungi, and bacteria, which is amounted to billions of tons Chitin is an insoluble N-acetylated glucosamine-containing polysaccharide,

whereas its deacetylation product, chitosan, generated naturally or by alkaline treatment

SC

becomes a positively charged polysaccharide. Chitosan is soluble in aqueous acidic solutions or in organic acids. Initially, solubility in acid is used to distinguish chitin from

M AN U

chitosan. Subsequently, the degree of acetylation (DA) is measured to distinguish the two in that chitin contains over 50% of N-acetylated glucosamine residues (DA > 50%), whereas chitosan has 0 to 50% N-acetylated glucosamine residues (0% ≤ DA < 50%). Interestingly when the DA is approaching 0%, chitosan is highly crystalline and becomes

TE D

insoluble again in acids. Therefore, DA measurement is a better and quantitative way to define specific chitins/chitosans compared to that of solubility. [2]

EP

In recent years, chitosan and its derivatives have attracted attention of many biologists due to newly discovered activities that shift its biological application from plants

[3]

to

AC C

animals, such as anti-microbial, anti-fungal, pro-coagulant, immune stimulating, hypolipidemia, and anti-tumor activities. [4, 5] For example, chitosan-based hemostatic products have been approved by the US food and drug administrations (FDA) and used clinically. [6, 7]

In addition, chitosan derivatives have been studied widely in non-viral gene- or

nanotechnology-based drug deliveries.

[8-13]

So far there are 34 completed or on-going

clinical trials in the US alone for chitin- and chitosan-based drugs. [14, 15]

3

ACCEPTED MANUSCRIPT Chitin and chitosan are different from most of clinical used drugs in that they are not pure compounds but a mixture of polysaccharides with variable DA, molecular weight distribution, and purity depending on the source of raw materials and methods used for

RI PT

preparations. [16, 17] At present it is largely unknown how the acetylated and deacetylated glucosamine residues distribute in the chitin/chitosan polysaccharide chains, an important structure feature related to their biological activities. Therefore, the DA and the

SC

distribution of acetylated glucosamine residues in chitin/chitosan are two important

M AN U

parameters highly desirable in defining the quality of chitin/chitosan for the purposes of medical applications.

Many assays for determination of DA in chitin/chitosan have been developed over the years, which can be classified into three categories: (1) chitin/chitosan free amine

TE D

measurement assays that include various types of titration, conductometry, potentiometry, ninhydrin, and pictric acid-based free amine measurement assays; [18-20] (2) destructive assays that include elemental analysis, acid- or enzyme-based hydrolysis followed by the

EP

DA measurement by colorimetric or high performance liquid chromatography (HPLC),

AC C

pyrolysis-gas chromatography, and thermal analysis using differential scanning calorimetry; [21-25] (3) spectrometry assays including UV, IR, 1H NMR, 13C NMR, and 15N NMR assays. [26-31]

The free amine measurement assays are only applicable to those chitins/chitosans soluble in specific solvents. In these assays, the measured free amine contents in intact chitin/chitosan polysaccharides are affected by the presence of impurities as well as by

4

ACCEPTED MANUSCRIPT final concentrations of chitin/chitosan, ionic strength of the solvent, pH, and temperature used in the assays. In the destructive assays, nitrogen content, released acetyl groups, or released glucosamine to acetyl glucosamine ratio are used to calculate the DA. Since

RI PT

many biological impurities are also composed of elemental C, H, O, and N and have acetyl groups, the DA measurement is largely affected by the presence of such impurities. In addition, quantitative destructive reactions are difficult to achieve when enzymes are

SC

used to release acetyl or glucosamine/N-acetylated glucosamine residues for such

M AN U

purposes. Furthermore, the assays in above two categories cannot provide with any structural information about how the acetylated and deacetylated glucosamine residues distribute in the polysaccharide chains of chitin/chitosan. 1

H NMR and UV assays are more sensitive than IR, 13C NMR, and 15N NMR in the DA

TE D

determination. [2] 1H NMR measures the DA by using the ratio of proton signal intensity of acetyl groups vs. a single or sum of proton intensities in chitin/chitosan whereas the UV assay measures the DA by quantifying the special absorbance provided by the double

EP

bond of NH-C=O in N-acetyl groups in chitin/chitosan at 200 nm or by first derivative

AC C

UV assay based on the similar principle. [32] It has been largely accepted that 1H NMR not only provides accurate measure of the DA but also offers partial information about how the acetylated and deacetylated glucosamine residues distributed in chitin/chitosan. However, recently published studies demonstrate that 1H NMR fails to detect up to 50% polysaccharides in chitosan and PEGylated chitosan copolymers

[33]

It has also been reported that even though the same

5

ACCEPTED MANUSCRIPT amount of oversulfated chondroitin sulfate (OSCS) and oversulfated dermatan sulfate (OSDS), two types of highly resembling chemically sulfated animal polysaccharides, possess the same numbers of acetyl groups, the 1H NMR proton signal intensities [34]

RI PT

reflecting the acetyl groups in OSDS are only half of that observed in OSCS.

Furthermore, 1H NMR could detect 0.1% OSCS in heparin based on the proton signal intensity of the acetyl groups in OSCS but could not detect 30% oversulfated heparan

SC

sulfate (OSHS) in heparin because the proton signal of the acetyl groups in OSHS is

M AN U

totally disappeared when mixed with heparin, [35, 36] which suggest that 1H NMR invisible or non-quantitative phenomena might be associated with polysaccharide structural analysis in general. Although 1H NMR-based chitin/chitosan analysis produces consistent spectra demonstrating great reproducibility towards polysaccharide analysis and has been

TE D

chosen as a standard method by the American Standard Test Method organization to determining the DA for chitosan,

[37]

the invisible and un-quantitative phenomena

observed in chitosan and other polysaccharides by 1H NMR analysis raise questions about

EP

the reliability of using 1H NMR analysis alone for chitin/chitosan DA measurement and

AC C

for comprehensive structure analysis. Therefore, a reliable method that can be used to measure simultaneously the DA as well as acetyl group distribution in chitin/chitosan for quality control of medically used chitins/chitosans is highly desirable.

Nitrous acid or HONO degradation assay, where HONO is the chemical formula of nitrous acid, has been well established for the structural analysis of heparin, an animal derived glucosamine-containing polysaccharide-based drug. [38-40] In the established 6

ACCEPTED MANUSCRIPT assays, deacetylated glucosamine residues can be cleaved quantitatively by HONO at pH 4.0 into 2, 5-anhydromannose (M). Furthermore, the newly generated aldehyde group in the M can be easily radiolabeled with NaB3H4. [38, 39] The radiolabeled

RI PT

compounds can be quantitatively analyzed, which have provided useful structural information for heparin and heparan sulfate during the past. [38, 39] Through literature

search, we found that the HONO assay has been used for chitin/chitosan degradation

SC

in several publications. [33, 41-43] Sashiwa et al reported that upon HONO deamination

M AN U

followed by NaBH4 reduction, most deamination products of various chitosans (over 50% of deacetylation) are oligomers of less than six units by gelfiltration separation and reflective index detection. These results suggest a random distribution of N-acetyl groups in the chitosans and potential use of this method for DA measurement.

TE D

In current study, we used HONO to cleave all un-acetylated glucosamine residues in chitosan quantitatively into 2, 5-anhydromannose (M) or M at reducing end of di-, tri-, and oligosaccharides. Instead of using radioactive material, a novel labeling strategy by

EP

using PMP, i.e. 1-phenyl-3-methyl-5-pyrazolone, to label all M residues to allow UV

AC C

quantification of each labeled M at 245 nm after C18 HPLC separation was developed. A formula was then used to calculate the DA based on the UV data.

2. Results

2.1. Optimizing HONO degradation and PMP labeling conditions for chitosan analysis.

7

ACCEPTED MANUSCRIPT A pre-requirement for chemical analysis of chitosan is to make it totally dissolved in proper solvents. We found that 2% acetic acid was not only served for the purpose but also had no deacetylation effects on chitosan after the solution was stored at room

RI PT

temperature for extended periods in comparison to that when strong acids, such as hydrochloric acid, was used as solvents. The dissolved chitosan in 2% acetic acid was then used for HONO treatment. The HONO treatment of chitosan in principle starts by

SC

nitrosation of the amino group of all the glucosamine residues followed by loss of N2 and

M AN U

ring contraction of the D-glucosamine residue simultaneously to form 2, 5anhydromannose (M) (Fig. 1A).We initially treated chitosan samples with HONO at pH 4 for 10 min, which is the established HONO reaction condition for heparin. However, the 2, 5-anhydromannose (M) generated by HONO treatment of chitosan could not be

TE D

monitored by conventional methods directly for lacking chromophore or fluorophore, we decided to develop a DA measurement protocol that is optimized for chitosan analysis. We have successfully used aniline-based reductive amination labeling strategy for

EP

glycosaminoglycan analysis during the past in which the amine groups in aniline react

AC C

with the aldehyde group in the sugar residues to form Schiff bases that are subsequently reduced by using sodium cyanoborohydride. [34, 44, 45] However, this strategy did not work for the HONO degraded chitosans (Data not shown). It has been demonstrated that PMP can be used for labeling amine-free or aminecontaining mono-, di-, tri, or oligosaccharides up to 19 mers as long as the saccharides contain an aldehyde group at the reducing end. [46] According to this protocol, two PMPs

8

ACCEPTED MANUSCRIPT react with one aldehyde group quantitatively to yield labeled compounds containing two PMPs per aldehyde group at the reducing end with UV absorbance at 245 nm (Fig. 1A). Since it has not been established if anhydromannose can be labeled by PMP, we first

RI PT

generated anhydromannose by treating the glucosamine hydrochloride with HONO at pH 4 and then performed PMP labeling by using the published reaction conditions.

[46]

After

analyzing the PMP labeled product by online UV detection after HPLC separation, we

SC

found that PMP labeling worked for the HONO generated anhydromannose. We then

M AN U

optimized the PMP labeling conditions by using different temperatures (Fig. 1B), different time points (Fig. 1C), and different anhydromannose to PMP ratios (Fig. 1D). The date in Fig. 1B, 1C, and 1D showed that the best PMP labeling condition was that the reactions were conducted at 70oC for 1 h with a 1:6 anhydromannose to PMP ratio. We

TE D

then used the optimized PMP labeling condition to test how long it was needed to complete chitosan degradation by HONO. The results of Fig. 1E would rather suggest that 15 min be the optimal time as a compromise between degradation and otherwise

EP

unspecific decomposition of the chitosan, as after a longer HONO incubation, the yield

AC C

seems to decrease again. Thus far, we established that the experimental conditions for both HONO degradation and PMP labeling. The schematic diagram of the strategy was summarized in the Scheme 1 where H, A, and M stand for glucosamine, Nacetylglucosamine, and 2, 5-anhydro-D-mannose, respectively. We also created a series of symbols in Scheme 1 to simplify the data presentation for subsequent analysis.

9

ACCEPTED MANUSCRIPT 2.2. Identifying mono-, di-, tri-, and oligosaccharides of HONO treated and PMP labeled chitosan 3 by MS analysis after C18 HPLC separation. To confirm the molecular identity of PMP labeled compounds, 1mg of chitosan 3 was

RI PT

treated with HONO for 15 min and followed by labeling with PMP at 70oC for 1 h. After removing excess PMP in the reaction mixture, a fraction of PMP labeled compounds were separated by HPLC using a C18 HPLC column and the total ion current (TIC)

−

(z1 491.19) molecular-ion matched the

M AN U

The major ion in peak 1 was a [M-H]

SC

chromatogram in negative ion mode was collected by the online MS (Fig. 2 A).

calculated molecular ion of M tagged with two PMPs. We then generated the M standard from the glucosamine hydrochloride by using the same procedure as for chitosan. Both elution time and molecular ion derived from HONO treated and PMP labeled M were

TE D

identical to that of the peak 1, which confirmed that peak 1 was M tagged with two PMPs. The major ion in peak 2 (z1 694.27) matched the molecular ion of the disaccharide AM tagged with two PMPs; the peak 3 (z1 897.34) was AAM tagged with two PMPs,

EP

Peak 4 (z1 1100.42) was AAAM tagged with two PMPs, etc. The peak 11 with [M-2H] 2−

AC C

(z2 1260.97) molecular-ion was also unambiguously identified as a 11mer (AAAAAAAAAAM) tagged with two PMPs. The MS data shown in Fig. 2 were listed in Table 1. As expected, when the observed UV peaks did not get baseline separation, the molecular ions from neighboring oligosaccharides were also observed (Fig. 2, lower panels from peaks 5 to 11). The MS data in both Fig. 2 and Table 1 showed that the HONO treatment plus PMP labeling produced chemically clean molecular ions,

10

ACCEPTED MANUSCRIPT indicating both HONO degradation and PMP labeling resulted in quantitative chemical reactions. The MS analysis confirmed the suitability of this assay for chitosan analysis.

RI PT

2.3. Online UV detection of HONO treated and PMP labeled chitosan products after C18 HPLC separation to determinate the DA and the distribution of acetyl groups in chitosan.

SC

Our goal was to develop a reliable UV-based assay for determination of the DA for chitosans. After optimizing the assay conditions (Fig. 1) and verifying the reliability of

M AN U

this assay by MS analysis (Fig. 2 and Table 1), we made another two chitosan samples with increased amount of N-acetylated glucosamine residues in chitosans (chitosan 2 and 3) (see “Experimental”). Three chitosan samples were treated with HONO and followed by labeling with PMP. After removing excess PMP in the reaction mixture, the PMP

TE D

labeled chitosan products were separated by HPLC using a C18 HPLC column and monitored at 245 nm by using an online UV detector. The UV data of the chitosan

EP

samples were shown in Fig. 3.

Based on the UV intensity of each peak area shown in Fig. 3A, 89% of glucosamine

AC C

(H) residues in chitosan 1 had HHH sequences since the red H would definitely produce anhydromannose (M) after HONO treatment. In contrast, 9% of glucosamine residues in chitosan 1 had HAHH sequences that produced the disaccharide AM after HONO treatment. Less than 2% glucosamine (H) existed in HAAHH sequences that produced the trisaccharide AAM after HONO treatment. Lacking long N-acetylglucosamine stretch,

11

ACCEPTED MANUSCRIPT such as AAA sequence in chitosan 1, indicated that the chemical deacetylating reaction might occur randomly Similarly, the chemical acetylating reactions increased the HAH sequences

RI PT

significantly in both chitosan 2 and chitosan 3, indicating the chemical acetylation reaction occurred randomly as well. The UV data in both Fig. 3B and 3C showed that the chemical acetylation reaction did not produce significant amount of N-acetylated

SC

glucosamine stretches in both chitosan 2 and chitosan 3, suggesting that the preexisting

M AN U

N-acetylglucosamine residues in the chitosan 1 did not promote or suppress the acetylating reactions of the neighboring glucosamine residues. The experimental data supported that increasing the amount of added acetic anhydride led to an increased acetylation of the chitosan 3 as expected.

TE D

In our current studies, HONO was used to cleave all glucosamine residues in chitosan into M or M at reducing end of di-, tri-, and oligosaccharides. PMP was then used to label all the Ms. Online UV detection allowed quantification of each M-containing mono-, di-,

EP

tri- and oligosaccharides at 245 nm after HPLC separation. The DA for chitosan samples

AC C

thus could be calculated based on the A/ (A+M) value derived from the UV data by using the following equation:

DA% =

2 + 23 + 34 + 45 ⋯ + 1011 1 + 22 + 33 + 44 + 55 ⋯ + 1111

Where the denominator was the sum of both glucosamine (H) and N-acetylated glucosamine (A) residues in chitosan, which equaled to the sum of the observed

12

ACCEPTED MANUSCRIPT absorbance intensity of each peak area multiplied by the total number of saccharides in each peak, i.e.P1+2P2+3P3+4P4+5P5+6P6+7P7+8P8+9P9+10P10+11P11. In parallel, the numerator was the sum of all N-acetylated glucosamine-containing of

degraded

chitosan,

P2+2P3+3P4+4P5+5P6+6P7+7P8+8P9+9P10+10P11.

which

equaled

to:

RI PT

peaks

Based on the UV data shown in Fig. 3, the DA was calculated for the three chitosan

2299 + 2 ∗ 549 ∗ 100% = 12.1% 21822 + 2 ∗ 2299 + 3 ∗ 549

M AN U

DA  % = DA  % =

SC

samples by using the equation:

7520 + 2 ∗ 4455 + 3 ∗ 1484 + 4 ∗ 638 + 5 ∗ 340 + 6 ∗ 156 ∗ 100% 49707 + 2 ∗ 7520 + 3 ∗ 4455 + 4 ∗ 1484 + 5 ∗ 638 + 6 ∗ 340 + 7 ∗ 156

DA  ! =

TE D

= 28.8%

45623 + 2 ∗ 18028 + 3 ∗ 2952 + 4 ∗ 951 + 5 ∗ 412 + 6 ∗ 212 + 7 ∗ 113 140141 + 2 ∗ 45623 + 3 ∗ 18028 + 4 ∗ 2952 + 5 ∗ 951 + 6 ∗ 412 + 7 ∗ 212 + 8 ∗ 113

EP

∗ 100% = 32.8%

The DA values for chitosan 1, 2, and 3 were 12.1%, 28.8%, and 32.8%, respectively,

AC C

based on the calculation.

2.4. The sensitivity and the reproducibility of the newly established method. To test applicability of the newly established method in general laboratory settings, we tested the sensitivity of the assay based on the dilution range of glucosamine standard by using a regular HPLC. In this experiment, glucosamine standard was HONO degraded

13

ACCEPTED MANUSCRIPT and PMP labeled. The HPLC separation and online UV detection results were plotted as peak area UV intensity against the nmol of PMP labeled M injected based on the initial glucosamine standard concentration (Fig. 4). The correlation between the two parameters

for PMP labeled compound was calculated to be 10 pmol.

RI PT

was linear with a correlation coefficient R2> 0.992. Based on the data, the detection limit

To test the reproducibility of the DA measurement by the newly established assay, the

SC

same experiment shown in Fig. 4 was repeated three times for each of the three chitosan

M AN U

samples by using regular HPLC/online UV analysis. The average DAs obtained for chitosans 1, 2, and 3 were 12.0±0.9%, 28.4±1.6%, and 32.3±1.8%, respectively, which was consistent with the results calculated based on the capillary HPLC data (Fig. 3).

2.5. DA of the chitosan 1 measured by a conductometric titration method.

TE D

This method is based on conductance measurement, which is a function of the sum of the conductance of each type of ions present in the chitosan solution. Since hydrogen and

EP

hydroxyl ions are the most conducting of all the ions, conductometric titration monitors the change in conductance due to changes of these ions in chitosan solution as a function

AC C

of added NaOH. In this assay, the conductivity is plotted against the volume of 0.1 M NaOH solution added. The following formula is used to calculate the DA: D A % = (1-

" # $% #&×(.((&∗&)& *+

×100 %

Where C is the molar concentration of NaOH used for the titration, V1 and V2 are the conductance transition points observed in the plot, 161 is the molecular weight of glucosamine, and Wc is the dry weight of chitosan used for the analysis.

14

ACCEPTED MANUSCRIPT One of three plots obtained by the conductometric titration measurement was shown in Fig. 5. The difference between V1 and V2 corresponded to the volume of 0.1 M NaOH required to neutralize the amino acid groups of glucosamine in the 200.0 mg of the

RI PT

chitosan 1 sample. As shown in Fig. 5, V2 was 21.5 mL whereas V1 was 10.5 mL. By putting these values into the formula, the DA value was 11.6%. Another two experiments had the same V1 value of 10.5 mL. However, the V2 values were 21 mL and 22 mL,

SC

which resulted in the DA values of 15.2% and 7.6%, respectively, based on the

DAexp1= (1-

(.& $&.,%&(.,×(.((&∗&)& (.$((

DA exp2= (1-

(.& $&%&(.,×(.((&∗&)&

DA exp3= (1-

(.& $$%&(.,×(.((&∗&)&

(.$((

(.$((

M AN U

calculations shown below. ×100 %=11.6%

×100 %=15.16% ×100 %=7.6%

measurements.

TE D

The average DA for the chitosan 1 was 11.6 ± 4% based on the three independent

The advantage of this method was that the DA was resulted in the direct measurement

EP

of exact amount of glucosamine residues present in the chitosan sample. Thus, the DA

AC C

values obtained were reliable. However, there were three major weaknesses of the method: 1. significant amount of chitosan sample (200.0 mg) was required for each measurement; 2. substantial standard deviation (SD: ± 4%) was observed as a result of the small difference in the observed V2 among three repeated experiments; and 3. this method did not provide any information about how N-acetylglucosamine residues distributed in the chitosan sample.

15

ACCEPTED MANUSCRIPT 2.6. DA of the chitosan 1 measured by direct 1H NMR method. The direct 1H NMR spectrum of chitosan 1 was shown in Fig. 6A. Theoretically, the DA of chitosan can be obtained using the formula below based on the 1H NMR data: -./0/0

RI PT

DA = -2345%//)* 100% Where the ICH3 is the integration of the three protons in the N-acetyl groups of glucosamine whereas the IGlcN-H is the integration of 6 protons in the H2, H3, H4, H5,

SC

and H6 positions in both glucosamine and N-acetyl glucosamine residues as indicated in

M AN U

Fig. 6B. Based on the proton integration data, ICH3 was 1.00 whereas the IGlcN-H was 30.16. Putting these values into the formula, the resulted DA value for the chitosan 1 was 6.6%. &/0

DA=0(.&)/) ∗ 100%=6.6%

TE D

The DA value obtained by direct 1H NMR method was significantly lower than that of the newly developed method as well as that of the conductometric titration method shown in Fig. 5. Considering the reported invisible issue associated with 1H NMR analysis of

EP

chitosans, [26, 33] we measured the DA of the same chitosan 1 sample using the indirect 1H

AC C

NMR method. [26]

2.7. DA of the chitosan 1 measured by indirect 1H NMR method. The assay principle of the indirect 1H NMR method is to release the 1H NMR invisible N-acetylglucosamine residues in chitosan by HONO treatment before the 1H NMR analysis as shown in Fig. 6D. The indirect 1H NMR spectrum of chitosan 1 was shown in

16

ACCEPTED MANUSCRIPT Fig. 6A. In the indirect 1H NMR method, the DA is calculated based on the following formula: DA% = A/(D+A) x 100%

RI PT

Where D = M1+ (H1+H2)/2 A = (A1+Ac/3)/2

M1, Ac, A1, H1, and H2 were labeled in red in Fig. 6D to represent specific protons in

SC

the HONO degraded chitosan.

M AN U

2.8. The comparison of repeatability, standard deviation (SD), and relative SD (RSD) of the DAs of chitosan 1 measured by four independent methods. The repeatability, standard deviation (SD), and relative SD (RSD) based on the data obtained from four different methods were calculated and summarized in Table 2.

TE D

Theoretically, conductometric titration is a reliable method for DA measurement. The average DA for chitosan 1 using this assay was 11.6±4% based on three independent

EP

measurements (Table 2). However, direct 1H NMR measurement of the chitosan 1 using the published protocol

[29]

produced a DA of 6.6%. Considering the reported invisible

AC C

issue associated with 1H NMR analysis of chitosan, chitosan 1 using the indirect 1H NMR method

[26, 33]

[26]

we measured the DA of the

where the chitosan 1 was first

degraded by using HONO followed by 1H NMR analysis. The indirect 1H NMR measurement produced a DA of 11.5%, which was in agreement with the results obtained by our (12.1±0.9%) and the conductometric titration (11.6±4%) measurements. Thus, we concluded that the newly developed method was more reliable than the direct 1H NMR

17

ACCEPTED MANUSCRIPT method. Our method reduced the SD and RSD significantly compared to that of the conductometric titration method (Table 2).

RI PT

3. Discussions and Conclusions

HONO degradation and PMP labeling approach for chitosan analysis was advantageous compared to other destruction methods described in the literature. In particular, those

Fmoc-OSu

(N-(9-fluorenylmethoxycarbonyloxy) [48]

phthalaldehyde (OPA)

succinimide)

[47]

or

o-

M AN U

either

SC

destruction methods involve in chitosan hydrolysis and glucosamine derivatization with

followed by HPLC-UV or fluorescence analysis in two

perspectives: 1. N-acetylglucosamine residues in chitosan were preserved during HONO degradation but totally destroyed during chitosan hydrolysis in the destruction methods.

TE D

Therefore, only the newly developed assay was capable to perform the analysis of both DA and the distribution of acetyl groups in chitosan whereas the destruction methods

EP

cannot perform these tasks. 2. Each anhydromannose (M) containing residue was tagged with two PMPs to produce a single UV peak that simplified mono-, di-, tri-, and

AC C

oligosaccharide quantification. In contrast, both Fmoc-OSu and OPA derivatization followed by HPLC-UV or fluorescence analysis produce two UV or fluorescence peaks of α- and β-enantiomers that make glucosamine quantification complicated. The chitosan sample (the chitosan 1) used in current study has high viscosity (100-200 mPa.s) as stated by the manufacturer, which might explain the discrepancy of the DA values obtained by the direct and indirect 1H NMR methods. These data suggest that

18

ACCEPTED MANUSCRIPT indirect 1H NMR method might be a better choice for the DA measurement of chitosan samples associated with high viscosity. In principle UV data was more reliable for quantifying PMP labeled saccharides than

RI PT

MS because UV absorbance of the labeled saccharides was not affected by the solvent used during or between different HPLC runs whereas the TIC chromatograms of MS was largely affected by the ionization conditions contributed by solvents, instrument, and

SC

other factors during, between, or even in the same run, which reflected the different peak

M AN U

intensities observed by the UV detection (Fig. 3C) or MS detection (Fig. 2A) for the same sample (chitosan 3).

Based on the 10 pmol detection limit derived from the data presented in Fig. 4, we reasoned that to detect an oligosaccharide with 10 N-acetylglucosamine repeats, i.e. the

TE D

11 mers as shown in Fig. 3, 21.9 ng of the 11 mers {10 pmol x (10 x 203 (MW of Nacetylglucosamine) + 1 x 161 (MW of glucosamine))} was needed for its detection by the regular HPLC whereas 42.2 ng of a 21 mers would be required for its detection. In our

EP

experiment, 1 mg of glucosamine or chitosan sample was weighed, HONO degraded,

AC C

PMP labeled, and ~1 µg of the labeled sample was used for capillary HPLC analysis whereas 100 µg of the labeled sample was used for regular HPLC analysis. By using the regular HPLC, even the sample had a DA of 100% with average 1000 Nacetylglucosamine residues per chitosan chain, 100 µg of injected sample should have 493 pmol PMP labeled products for UV detection, which was 49 times above the detection limit. Thus, this assay was sensitive enough for both chitin and chitosan

19

ACCEPTED MANUSCRIPT analysis since the availability of chitin/chitosan samples had never been a limiting factor for any kinds of chemical analysis. The sensitivity, reproducibility, and reliability of the DA values obtained by the newly

RI PT

established method were better than that of the conductometric titration and direct 1H NMR methods. However, since we did not have a titration burette available that allows at least a 10-20 more accurate measurement by the titration assay, which would lead to a

SC

much smaller error.

M AN U

Thus, we hope this method would be useful not only for basic chitin/chitosan research but also for quality control of medically used chitosan and derivatives in near future by any laboratory that is equipped with a HPLC and has an online UV detector. In summary, a HONO degradation plus PMP labeling followed by HPLC separation

TE D

and online UV detection assay was developed for measuring the DA in chitosans. The reliability of the assay was confirmed by online MS analysis where 11 different kinds of mono-, di-, tri-, and oligosaccharides were identified unambiguously. Moreover, by using

EP

this assay, we showed that acetylated and deacetylated glucosamine residues distributed

AC C

in three related chitosan samples in a random fashion, suggesting both deacetylation and acetylation reaction occurred randomly where the acetylated or free glucosamine residues in chitosan did not promote or suppress acetylating or deacetylating reactions of their neighboring sugar residues. Therefore, this newly developed method was not only useful for measuring the DA but also provided useful structural information as well as the mechanisms for chemical acetylation and deacetylation reactions of chitins/chitosans.

20

ACCEPTED MANUSCRIPT 4. Experimental

4.1. Materials and experimental procedures.

RI PT

Chitosan was purchased from Aladdin Inc., China, which is a white powder obtained by chemical deacetylation of chitin using alkaline treatment. The chitosan was oven-dried at 70oC for 4 h before use and named chitosan 1 in this manuscript. Glucosamine chloride

SC

was purchased from Sigma, USA. Acetic anhydride and nitrous acid were obtained from Sinopharm Chemical Reagent Co, Ltd., China. PMP and glucosamine chloride were

M AN U

purchased from Sigma Aldrich, St. Louis, MO.

The experimental conditions were extensively optimized and the optimized experimental procedures were described below.

TE D

4.2. HONO degradation.

Glucosamine chloride (1 mg) or chitosan samples (1 mg) were accurately weighed and

EP

then dissolved in 2% acetic acid to adjust pH to 4. Equal volumes of HONO solution (5.5 M NaNO2 and 1M H2SO4 with 5:2 volume ratio with a final pH adjusted to 4.0) was

AC C

added. After 15 min, the reaction was terminated by adjusting the pH to 7 using 22% NH4OH and ready for the PMP labeling.

4.3. PMP labeling.

The HONO treated samples were used for PMP labeling directly by adjusting to pH ≥ 10 and then by adding 75 µL of 0.5 M PMP dissolved in methanol. The labeling reaction was conducted at 70 oC for 1 h and then cooled to room temperature followed by 21

ACCEPTED MANUSCRIPT neutralization using acetic acid. The extra PMP in the labeling reaction was extracted by using 1:2.5 volumes of chloroform three times as previously described. [46]

4.4. Chemical acetylation to obtain chitosan 2 and chitosan 3. [49]

RI PT

A previously established chitosan acetylation method was used with modifications.

In brief, 10 mg of the oven dried commercial chitosan sample (chitosan 1) was

SC

suspended in water (0.2 mL) for 1 h followed by adding 20% aqueous acetic acid (0.3 mL) to the suspension to dissolve chitosan. Different amounts of acetic anhydride was

M AN U

dissolved in 0.5 mL of methanol and then added to the chitosan solution. The chitosan solution immediately became a gel when the acetic anhydride solution was added. After stirring for 18 h at room temperature, the N-acetylated chitosan gel was washed with equal volumes of water/acetone (1:7 by weight) three times. The chemically acetylated

TE D

chitosans were then freeze-dried and ready for further analysis. When using 0.2 mol acetic anhydride towards 1 mol free amine in the chitosan 1 for

EP

the acetylating reaction, the product obtained was named chitosan 2 whereas when using 0.5 mol acetic anhydride towards 1 mol free amine in the chitosan 1 for the acetylating

AC C

reaction, the obtained product was named chitosan 3.

4.5. Liquid Chromatography/Mass Spectrometry analysis. The HONO degraded and the PMP labeled samples were separated using an Agilent 1200 Infinity capillary liquid chromatography system with a ZORBAX 300SB-C18 chromatogram column (0.5x250mm). Solvent A was 10 mM ammonium acetate adjusted

22

ACCEPTED MANUSCRIPT to pH 5.0 by using acetic acid and solvent B was 100% acetonitrile. The gradient used was 20-25% B in 40 min and 25% B over 30 min. The column was then washed with 50% B for 10 min and equilibrated with 20% solvent B. The flow rate was 10 µL/min.

RI PT

The UV signal was detected by measuring absorption at 245 nm of the PMP labeled samples eluted from a C18 column. MS analysis was performed by coupling the HPLC with a thermo LTQ-XL mass spectrometer as previously described. [50] Total ion current

SC

(TIC) chromatograms in negative ion mode were collected by scanning the m/z ranged

M AN U

from 300–2000. The mass chromatograms were processed with Thermo Xcalibur software.

To test the reproducibility of the newly established method for DA measurement, chitosans 1, 2, and 3 were analyzed by HONO degradation, PMP labeling, and regular

TE D

HPLC/UV analysis three times by using an Agilent 1260 HPLC with the same solvent system as described for the capillary HPLC.

EP

4.6. Determination of DA by conductometric titration method. The same oven-dried chitosan sample (chitosan 1, 200.0 mg) was dissolved in 0.1 M

AC C

HCl to make an acidic chitosan solution in a total volume of 200.0 mL. Half mL of a 0.1 M NaOH solution was titrated into the acidic chitosan solution every 30 seconds and the conductivity was recorded every 30 seconds as well based on the published method. The experiment was repeated three times.

4.7. Direct 1H-NMR method for DA measurement.

23

[19]

ACCEPTED MANUSCRIPT The same oven-dried chitosan sample (chitosan 1, 10 mg) was D2O exchanged 3 times and then dissolved in 0.2 M CF3COOD/D2O solution. The 1H NMR spectrum was recorded using Varian 400M NMR spectrometer. The DA was calculated based on the 1H

4.8. Indirect 1H-NMR method for DA measurement.

RI PT

NMR data obtained according to the published method. [26]

SC

The principle of the indirect 1H NMR method is to release the 1H NMR invisible Nacetylglucosamine residues in chitosan by HONO treatment before the 1H NMR analysis

M AN U

according to the published report. [33] To perform the indirect 1H NMR method, the same oven-dried chitosan sample (chitosan 1, 10 mg) was D2O exchanged 3 times and then dissolved in 0.2 M CF3COOD/D2O solution followed by adding 1 mg NaNO2 in D2O. 1H NMR spectrum was recorded by using Varian 400 M NMR spectrometer. The DA was

TE D

calculated based on the 1H NMR data obtained. [26]

EP

4.9. Statistical Analysis

All data are presented as the mean ± SD. SD and relative SD (RSD) were calculated

AC C

by using the Graph Pad Software.

Acknowledgements

This study was supported in part by Natural Science Foundation of China (Grant No. 91129706) and the NSFC-Shandong Joint Fund for Marine Science Research Centers (Grant No. U1406402).

24

ACCEPTED MANUSCRIPT Abbreviations DA, degree of acetylation; PMP, 1-phenyl-3-methyl-5- pyrazolone; HPLC, high performance liquid chromatography; MS, mass spectrometry; HONO, nitrous acid; M,

References:

RI PT

anhydromannose; A, N-acetylglucosamine.

Yeul, V.S., Rayalu, S.S., J Polym Environ,2013 21 606-614.

[2]

Kasaai, M.R., J Agr Food Chem,2009 57 1667-1676.

[3]

Cota-Arriola, O., Cortez-Rocha, M.O., Burgos-Hernandez, A., Ezquerra-Brauer, J.M.,

SC

[1]

Plascencia-Jatomea, M., J Sci Food Agric,2013 93 1525-1536. [4]

Jarmila, V., Vavrikova, E., Curr Pharm Des,2011 17 3596-3607.

[5]

Stefan, J., Lorkowska-Zawicka, B., Kaminski, K., Szczubialka, K., Nowakowska, M., Korbut,

M AN U

R., J Physiol Pharmacol,2014 65 341-347. [6]

Jayakumar, R., Prabaharan, M., Sudheesh, K.P., Nair, S.V., Tamura, H., Biotechnol Adv,2011

[7]

Brown, M.A., Daya, M.R., Worley, J.A., J Emerg Med,2009 37 1-7.

[8]

Singh, B., Choi, Y.J., Park, I.K., Akaike, T., Cho, C.S., J Nanosci Nanotechnol,2014 14 564-

29 322-337.

576. [9]

Saranya, N., Moorthi, A., Saravanan, S., Devi, M.P., Selvamurugan, N., Int J Biol

TE D

Macromol,2011 48 234-238.

Ragelle, H.S., Vandermeulen, G.L., Préat, V., J Control Release,2013 172 207-218.

[11]

Ding, F., Deng, H., Du Y, Shi, X., Wang, Q., Nanoscale,2014 6 9477-9493.

[12]

Amoozgar, Z., Park, J., Lin, Q., Yeo, Y., Mol Pharm,2012 9 1262-1270.

[13]

Raftery, R., O'Brien, F., Cryan, S., Molecules,2013 18 5611-5647.

[14]

Gov, C. 29 studies found for chitosan in ClinicalTrials.gov., 2014.

[15]

Gov, C. 6 studies found for chitin in ClinicalTrials.gov., 2014.

[16]

Tolaimate, A., Desbrières, J., Rhazi, M., Alagui, A., Vincendon, M., Vottero, P.,

EP

[10]

AC C

Polymer,2000 41 2463-2469. [17]

Pacheco, N., Garnica-Gonzalez, M., Gimeno, M., Barzana, E., Trombotto, S., David, L.,

Shirai, K., Biomacromolecules,2011 12 3285-3290.

[18]

Jiang, X., Chen, L., Zhong, W., Carbohyd Polym,2003 54 457-463.

[19]

Raymond, L., Morin, F.G., Marchessault, R.H., Carbohyd Res,1993 246 331-336.

[20]

T Ei, K., Kohara, T., Anal Chim Acta,1976 83 59-65.

[21]

Sato, H., Mizutani, S., Tsuge, S., Ohtani, H., Aoi, K., Takasu, A., Okada, M., Kobayashi, S.,

[22]

Nanjo, F., Katsumi, R., Sakai, K., Anal Biochem,1991 193 164-167.

[23]

Kittur, F.S., Harish Prashanth, K.V., Udaya Sankar, K., Tharanathan, R.N., Carbohyd

Kiyosada, T., Shoda, S., Anal Chem,1998 70 7-12.

Polym,2002 49 185-193. [24]

Holan, Z., Votruba, J., Vlasáková, V., J Chromatogr A,1980 190 67-76.

25

ACCEPTED MANUSCRIPT [25]

Niola, F., Basora, N., Chornet, E., Vidal, P.F.O., Carbohyd Res,1993 238 1-9.

[26]

Varum, K.M., Anthonsen, M.W., Grasdalen, H., Smidsrod, O., Carbohydr Res,1991 211 1723.

[27]

Kasaai, M.R., Carbohyd Polym,2010 79 801-810.

[28]

Heux,

L.,

Brugnerotto,

J.,

Desbrieres,

J.,

Versali,

M.F.,

Rinaudo,

M.,

Biomacromolecules,2000 1 746-751. Lavertu, M., Xia, Z., Serreqi, A.N., Berrada, M., Rodrigues, A., Wang, D., Buschmann, M.D., Gupta, A., J Pharm Biomed Anal,2003 32 1149-1158.

RI PT

[29]

Guinesi, L.S., Cavalheiro, É.T.G., Thermochim Acta,2006 444 128-133.

[31]

KASAAI, M., Carbohyd Polym,2008 71 497-508.

[32]

Wu, T., Zivanovic, S., Carbohyd Polym,2008 73 248-253.

[33]

Novoa-Carballal, R., Riguera, R., Fernandez-Megia, E., Mol Pharm,2013 10 3225-3231.

[34]

Pan, J., Qian, Y., Zhou, X., Lu, H., Ramacciotti, E., Zhang, L., J Biol Chem,2010 285 22966-

SC

[30]

22975. [35]

Pan, J., Qian, Y., Zhou, X., Pazandak, A., Frazier, S.B., Weiser, P., Lu, H., Zhang, L., Nat Biotechnol,2010 28 203-207, 207-211.

Keire, D.A., Mans, D.J., Ye, H., Kolinski, R.E., Buhse, L.F., J Pharm Biomed Anal,2010 52 656-664.

[37]

M AN U

[36]

Method, A.S.T. Method for determining degree of deacetylation in chitosan salts by proton nuclear magnetic resonance (1H NMR) spectroscopy., 2003, pp. F2260-F2263.

[38]

Shively, J.E., Conrad, H.E., Biochemistry-Us,1976 15 3932-3942.

[39]

Zhang, L., Lawrence, R., Schwartz, J.J., Bai, X., Wei, G., Esko, J.D., Rosenberg, R.D., J Biol Chem,2001 276 28806-28813.

Gill, V.L., Wang, Q., Shi, X., Zaia, J., Anal Chem,2012 84 7539-7546.

[41]

Tommeraas, K., Varum, K.M., Christensen, B.E., Smidsrod, O., Carbohydr Res,2001 333 137-144.

TE D

[40]

[42]

Allan, G.G., Peyron, M., Carbohydr Res,1995 277 257-272.

[43]

Sashiwa, H., Saimoto, H., Shigemasa, Y., Tokura, S., Carbohydr Res,1993 242 167-172. Dündar, M., Müller, T., Zhang, Q., Pan, J., Steinmann, B., Vodopiutz, J., Gruber, R., Sonoda,

EP

[44]

T., Krabichler, B., Utermann, G., Baenziger, J.U., Zhang, L., Janecke, A.R., The American Journal of Human Genetics,2009 85 873-882. Pan, J., Qian, Y., Zhou, X., Pazandak, A., Frazier, S.B., Weiser, P., Lu, H., Zhang, L.,

AC C

[45]

Glycobiol Insights,2010 2010 1-12.

[46]

Honda, S., Akao, E., Suzuki, S., Okuda, M., Kakehi, K., Nakamura, J., Anal Biochem,1989 180 351-357.

[47]

Yan, X., Evenocheck, H.M., Carbohyd Polym,2012 87 1774-1778.

[48]

Studelska, D.R., Giljum, K., McDowell, L.M., Zhang, L., Glycobiology,2006 16 65-72.

[49]

Hirano, S., Ohe, Y., Ono, H., Carbohydr Res,1976 47 315-320.

[50]

Zeng, Y., Han, Z., Qiu, P., Zhou, Z., Tang, Y., Zhao, Y., Zheng, S., Xu, C., Zhang, X., Yin, P., Jiang, X., Lu, H., Yu, G., Zhang, L., PLoS One,2014 9 e103880.

26

ACCEPTED MANUSCRIPT Table 1. A summary of detailed PMP labeled chitosan mono-, di-, tri-, and oligosaccharide MS data as shown in Fig. 2.

1 2 3

2,5-anhydromannose Disaccharide Trisaccharide

M AM AAM

4

Tetrasaccharide

AAAM

1101.44

5

Pentasaccharide

AAAAM

1304.52

6

Hexasaccharide

AAAAAM

1507.60

7

Heptasaccharide

AAAAAAM

1710.68

8

Octosaccharide

AAAAAAAM

1913.76

Nonasaccharide

AAAAAAAAM

2116.84

10 11

Decasaccharide Undecasaccharide

AAAAAAAAAM AAAAAAAAAAM

2319.92 2523.00

1100.42/549.71

TE D EP AC C 27

44

1303.50/651.25

41

none/752.78

41

none/854.32

39

none/955.86

38

none/1057.89

M AN U

9

RT (min) 67 51 46

RI PT

M/Z (Tagged with two PMPs) Theoretical Value (z1) Observed (z1/z2) 492.20 491.19 695.28 694.27 898.36 897.34

SC

Structure Code

HONO degraded Chitosan

none/1159.46 none/1260.97

37

36 35

ACCEPTED MANUSCRIPT Table 2. The DA values, repeatability, standard deviation (SD), and relative SD (RSD) of

Experiment

DA(%)

HONO/PMP/HPLC

1

11.2

2

12.1

3

13.0

1

11.6

2

15.6

3

7.6

Conductometric Titration

1

SD(%)

RSD(%)

0.9

7.5

4.0

34

H NMR

1

6.6

N/A

N/A

1

1

11.5

N/A

N/A

AC C

EP

TE D

M AN U

HONO/ H NMR

SC

Methods

RI PT

the chitosan 1 measured by different methods.

28

ACCEPTED MANUSCRIPT

Figure captions Scheme 1. Schematic diagram of HONO degradation and PMP labeling used for chitosan

RI PT

analysis. The chemical structures and symbols used for chitosan and degraded product presentations in this scheme were used for data presentations and explanations elsewhere in this manuscript.

Figure 1. Optimizing HONO degradation and PMP labeling conditions for chitosan analysis.

SC

Chitosan was treated with HONO and followed by labeling with PMP. After removing excess PMP in the reaction mixture, the PMP labeled products were separated by HPLC using a C18 HPLC

M AN U

column and the PMP labeled products were monitored at 245 nm by using an online UV (see “Experimental”) and the maximum UV absorbance of PMP labeled anhydromannose (M) from three independent analyses at each data point were plotted. A. schematic diagram of HONO degradation and PMP labeling for chitosan analysis. B. Optimizing reaction temperature for PMP labeling. C. Optimizing reaction time for PMP labeling. D. Optimizing the amount of PMP needed to achieve

TE D

maximum labeling. E. Optimizing reaction time for HONO treatment.

Figure 2. MS analysis of the HONO treated, PMP labeled, and LC separated chitosan 3 mono-, di-, tri, and oligosaccharides. Chitosan 3 was treated with HONO and followed by labeling with

EP

PMP. After removing excess PMP in the reaction mixture, the PMP labeled compounds were

AC C

separated by LC using a C18 column. The total ion current (TIC) was collected by using an online mass spectrometer (see “Experimental”). A: TIC chromatogram of the MS analysis; B: the accumulated m/z for each peak. Figure 3. UV profiles of three HONO treated and PMP labeled chitosan samples after LC separation. Three related chitosan samples were treated with HONO and followed by labeling with PMP. After removing excess PMP in the reaction mixture, the PMP labeled chitosan products were separated by regular HPLC using a C18 HPLC column and the UV absorbance was monitored at 245

29

ACCEPTED MANUSCRIPT

nm by using an online detector (see “Experimental”). A. Chitosan 1, a commercial chitosan sample, i.e. a chemically deacetylated chitin; B. Chitosan 2, an acetylated chitosan when 0.2 mol acetic anhydride towards 1 mol free amine in chitosan 1 was use in the acetylating reaction; C. Chitosan 3,

RI PT

an acetylated chitosan when 0.5 mol acetic anhydride towards 1 mol free amine in chitosan 1 was use in the acetylating reaction (see “Experimental”).

Figure 4. Linear correlation between UV detection to nmol of PMP labeled M injected based on

SC

initial glucosamine standard concentration used for HONO degradation and PMP labeling. In this experiment, glucosamine standard was HONO degraded and PMP labeled. A regular HPLC

M AN U

(Agilent HPLC1260) with online UV detector was used for data collection.

Figure 5. Conductometric titration curve of the chitosan 1. The chitosan 1 (200.0 mg) was dissolved in 0.1 M HCl to make an acidic chitosan solution in a total volume of 200.0 mL. Half mLof a 0.1 M NaOH solution was titrated into the acidic chitosan solution every 30 seconds and the

TE D

conductivity was recorded every 30 seconds as well based on the published method [19]. The experiment was repeated three times. One of the three conductometric titration curves was shown (see “Experimental”).

EP

Figure 6. Direct and indirect 1H NMR spectra of the chitosan 1. (A). Direct 1H NMR spectra

1

AC C

of the chitosan 1(see “Experimental”). (B). The protons used for calculating the DA for the direct H NMR method. In this method, the DA was calculated based on the following formula: DA =

(ICH3/3)/(IGlcN-H/6)* 100%, where the ICH3 is the integration of the three protons in the Nacetyl groups of glucosamine whereas the IGlcN-H is the integration of 6 protons in the H2, H3, H4, H5, and H6 positions in both glucosamine and N-acetyl glucosamine residues. We labeled ICH3 protons in green and IGlcN-H protons in red to indicate the protons in the chitosan 1 that were used for DA calculation. (C). Indirect

30

1

H NMR spectra of the chitosan 1 (see

ACCEPTED MANUSCRIPT

“Experimental”). (D). The protons used for calculating the DA for the indirect 1H NMR method. In this method, the DA was calculated based on the following formula: DA% = A/(D+A) x 100%, where D = M1+ (H1+H2)/2 and A = (A1+Ac/3)/2. We labeled all the protons including

RI PT

M1, Ac, A1, H1, and H2 in red to indicate the protons in the HONO degraded chitosan 1 that

AC C

EP

TE D

M AN U

SC

were used for DA calculation.

31

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

RI PT

Graphical abstract

Highlights

A novel method for determination of the degree of acetylation (DA) in chitosan

SC

•

was developed.

The distribution of acetyl groups and acetylation/deacetylation mechanisms of chitosan were also revealed.

M AN U

•

•

The reliability of the method was confirmed by online LC-MS analysis.

•

The accuracy of the DA measurement was improved compared to established

This method could be used by any laboratory equipped with a HPLC and an

EP

online UV detector.

AC C

•

TE D

methods.