Mechanism of interaction between hydroxypropyl cellulose and water in aqueous solutions: Importance of polymer chain length

ORIGINAL ARTICLE
Mechanism o in e ac ion be ween hyd oxyp opyl cellulose
and wa e in aqueous solu ions: Impo ance o polyme
chain leng h
Manuel Ma in-Pas o
1
| Edmon S oyano
2
1
Unidad de Resonancia Magné ica,
Uni e si y o San iago de Compos ela,
RIAIDT, San iago de Compos ela, Spain
2
Nisso Chemical Eu ope, Duesseldo ,
Ge many
Co espondence
Edmon S oyano , Nisso Chemical
Eu ope, Be line Allee 42, 40212
Duesseldo , Ge many.
Email: [email p o ec ed]
Abs ac
The u iliza ion o hyd oxyp opyl cellulose (HPC) can be ega ded as unex-
pec ed wi h ega d o ce ain applica ions, such as being employed as a solu-
bili y enhance o poo ly soluble d ugs and as a solubilizing agen o
nano-suspensions and amo phous solid dispe sions. Howe e , he bes esul s
we e ob ained o low-molecula weigh (M
w
) HPC g ades wi h a sho -chain
s uc u e. The e o e, in his s udy, se en g ades o HPC wi h di e en poly-
me chain leng hs (M
w
) a e analyzed in a ious aqueous solu ions by a combi-
na ion o
1
H quan i a i e NMR spec oscopy, di usion NMR spec oscopy,
and wa e ligand obse ed ia g adien spec oscopy; hese in es iga ions p o-
ide insigh s in o he ema kable solubilizing p ope y o HPC a he molecu-
la and sup amolecula le els. Fu he mo e, he hyd a ion and he wa e
esidence ime a e ound o be s ongly dependen on he polyme chain
leng h o HPC. The quan i a i e esul s ob ained he ein indica e ha HPCs
wi h sho e chain leng hs e ain smalle amoun s o wa e a ound hei
hyd a ed molecules, as compa ed o hei coun e pa s wi h longe chain
leng hs.
KEYWORDS
1
H wa e ligand obse ed ia g adien spec oscopy (Wa e LOGSY), di usion-o ien ed
spec oscopy (DOSY), hyd oxyp opyl cellulose, quan i a i e pe ec -echo Wa e ga e
1
H-NMR,
a iable empe a u e NMR
1|INTRODUCTION
Hyd oxyp opyl cellulose (HPC) is a cellulose e he o en
used in pha maceu icals as a able binde and ilm-coa ing
agen . Recen s udies ha e demons a ed new and unex-
pec ed ea u es o hese polyme s as solubili y enhance s
o poo ly soluble d ugs as well as o solubilizing nano-
suspensions and amo phous solid dispe sions.
[1–3]
These
s udies e ealed ha he bes esul s could be achie ed
wi h low-molecula weigh (M
w
) polyme s co esponding
o sho -chain HPC g ades.
The chemical s uc u e o HPC has been in es iga ed in
he pas by
1
Hand
13
C NMR spec oscopies.
[4,5]
Kimu a
e al.
[4]
elucida ed he deg ee o subs i u ion, molecula
subs i u ion, and he di e en eac i i ies o he hyd oxyl
g oups in cellulose molecules based on he
13
C-NMR peak
in eg als o samples p epa ed in D
2
O. Using he same ca -
bon NMR me hod, Desai e al.
[5]
analyzed he pe o mance
Recei ed: 20 Ma ch 2020 Re ised: 8 Ap il 2020 Accep ed: 9 Ap il 2020
DOI: 10.1002/pol.20200185
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and ep oduc ion in any medium, p o ided
he o iginal wo k is p ope ly ci ed.
© 2020 The Au ho s. Jou nal o Polyme Science published by Wiley Pe iodicals LLC.
1632 J Polym Sci. 2020;58:1632–1641.wileyonlinelib a y.com/jou nal/pol
o HPCs comp ising di e en cloud poin s, wi h an empha-
sis on he a io o ou e /single ca bons e sus inne ca -
bons. These undamen al wo ks assumed ha he HPC
molecula s uc u e is independen o he polyme chain
leng h and ha he examined g ades in hese s udies a e
ep esen a i e o he en i e HPC amily.
Howe e , he easons o he be e solubilizing cha ac-
e is ics o low M
w
HPC g ades o poo ly soluble d ugs
ha e no been clea ly unde s ood ye . The e a e no mean-
ing ul di e ences in he physicochemical p ope ies (solu-
bili y, dynamic ligh sca e ing, mass spec ome y, and
inne /ou e ca bon a io) o di e en g ades o his cellu-
losic e he . To ob ain meaning ul insigh s on he HPC
beha io in aqueous solu ions and he e ec o molecula
weigh a he molecula and sup amolecula le els, we
employed NMR echniques o quan i a i ely s udy he sel -
agg ega ion p ope ies o HPC in solu ion and he accessi-
bili y o wa e in di e en pa s o he polyme s uc u e.
To comp ehensi ely s udy hese e ec s, we analyze and
compa e he esul s ob ained o se en di e en HPC
g ades wi h di e en molecula weigh s (chain leng hs).
2|EXPERIMENTAL METHODS
2.1 |Chemicals
Se en HPC g ades wi h a e age molecula weigh s o
2,500,000 (HPC-VH), 1,000,000 (HPC-H), 700,000 (HPC-
M), 140,000 (HPC-L), 100,000 (HPC-SL), 40,000 (HPC-
SSL), and 20,000 (HPC-UL) we e p ocu ed om Nippon
Soda (Japan), o his s udy.
2.2 |Expe imen s
NMR expe imen s we e conduc ed on a B uke NEO 17.6 T
spec ome e (p o on esonance a 750 MHz) equipped
wi h a
1
H/
13
C/
15
N iple esonancep obeandashielded
PFG z-g adien . TopSpin 4.0 was used as he con ol so -
wa e. All he spec a we e p ocessed using Mes eNo a
12.0 (Mes elab Resea ch, Inc.). Chemical shi s we e
e e enced au oma ically agains a deu e ium lock. All spec-
a we e eco ded a 25C unless o he wise indica ed.
2.2.1 |Sample p epa a ion
Samples o HPC-VH, HPC-H, HPC-M, HPC-L, HPC-SL,
HPC-SSL, and HPC-UL we e p epa ed a mol/mol con-
cen a ion o 1.8% in D
2
O (D 99.9%) (Eu iso op, Inc.).
Subsequen ly, 500 μl o he samples we e ans e ed o
he 5 mm s anda d NMR ubes. Fo eco ding he wa e
ligand obse ed ia g adien spec oscopy (Wa e LOGSY)
spec a,
[6]
simila sample concen a ions and olumes
we e used. Unless o he wise indica ed, he NMR spec a
discussed in his s udy e e o he samples p epa ed
in D
2
O.
Samples o HPC-VH, HPC-H, HPC-M, HPC-L, HPC-
SL, HPC-SSL, and HPC-UL we e p epa ed a mol/mol
concen a ion o 1.8% in D
2
O (D 99.9%) (Eu iso op, Inc.).
Subsequen ly, 500 μl o he samples we e ans e ed o
he 5 mm s anda d NMR ubes. Fo eco ding he wa e
ligand obse ed ia g adien spec oscopy (Wa e LOGSY)
spec a,
[6]
simila sample concen a ions and olumes
we e used. Unless o he wise indica ed, he NMR spec a
discussed in his s udy e e o he samples p epa ed
in D
2
O.
2.2.2 |Va iable empe a u e
quan i a i e
1
H spec a
P o on quan i a i e monodimensional spec a (
1
H) o he
se en p epa ed HPC samples we e eco ded a 25 and
40C. Each spec um was eco ded o e 128 scans wi h a
pulse-acquisi ion sequence using an in e scan delay (d
1
)o
10 s and acquisi ion ime (aq) o 2.75 s. Each spec um
was p ocessed and analyzed by in eg a ion o he a ea
co esponding o alipha ic p o ons (δ om 0.5 o 1.3 ppm),
py anose ing p o ons (δ om 2.88 o 4.17 ppm), and a
esidual wa e peak (HDO) a ~4.7 ppm in he spec um
measu ed a 25C and 4.5 ppm a 40C. In each spec um,
peak a eas we e no malized wi h espec o he in eg al o
he HDO peak; he signal (100%) enhancemen ac o was
calcula ed independen ly o he peaks o py anose ings
and alipha ic chains using Equa ion (1),
Enh_A = In _A40C=In _HDO40C

=In _A25C=In _HDO25C

,
ð1Þ
whe e In _A
40C
and In _A
25C
e e o he in eg als o
peak A in he NMR spec a measu ed a 40C and 25C,
espec i ely. In his case, he wo peaks o in e es co -
esponded wi h he py anose ing p o ons and alipha ic
p o ons. In _HDO
40C
and In _HDO
25C
e e o he no -
malized in eg als o HDO in he NMR spec a eco ded a
40C and 25C, espec i ely.
2.2.3 |Di usion NMR spec um
DOSY (pulse sequence ledbpgp2s o he B uke lib a y)
[7]
analysis o he se en chosen HPC samples was conduc ed
wi h he BiPola -G adien s-Longi udinal EDdy Cu en
MARTIN-PASTOR AND STOYANOV 1633
Compensa ed-S imula ed-Echo (BPP-LED-STE) expe i-
men s. G adien pulses encoding di usion we e a ied
linea ly om 1 o 50 G/cm along 32 poin s in he di u-
sion dimension. The du a ion o each pai o bipola g a-
dien s δwas 8 ms and he di usion ime δwas 300 ms.
Spec a we e acqui ed wi h a elaxa ion delay (d
1
)o 3s,
acquisi ion ime o 2.75 s, and 16 scans a each poin in
he di usion dimension. A e Fou ie ans o ma ion in
he
1
H dimension, se e al peaks we e selec ed o analyz-
ing sel -di usion coe icien s. In each case, peak in ensi y
along he di usion dimension was i ed o he mono-
exponen ial S ejskal-Tanne equa ion by using he O igin
8.0 so wa e o de e mine he sel -di usion coe i-
cien (D).
2.2.4 |Quan i a i e
1
H pe ec -echo
spec a
One-dimensional
1
H quan i a i e pe ec -echo Wa e ga e
spec a (
1
H quan i a i e)
[8]
we e eco ded o e 128 scans
o he se en samples p epa ed in H
2
O as he sol en .
The in e scan elaxa ion delay (d
1
) was 10 s while he
acquisi ion ime (aq) was 1.38 s.
2.2.5 |Wa e LOGSY spec a
One-dimensional Wa e LOGSY
[6]
spec a (pulse
sequence Wa e LOGSY o he B uke lib a y) we e
eco ded o he se en chosen HPC samples dissol ed in
H
2
O. A 180in e sion pulse was applied o e he H
2
O
signal (~4.7 ppm) using a sinc-shaped selec i e pulse o
7.5 ms and bandwid h o 75 Hz. The s ong wa e peak
was supp essed be o e de ec ion by using a pai o
exci a ion-sculp ing echoes based on a selec i e 180
Gaussian shaped e ocusing pulse o 2 ms wi h PFG g a-
dien s o 1 ms du a ion and powe le els o 31 and
11 G/cm. The spec a we e eco ded o e 256 scans wi h
in e scan elaxa ion delay (d
1
) and acquisi ion imes
(aq) o 4 s and 1.38 s, espec i ely. Fu he mo e, Wa e -
LOGSY spec a we e acqui ed o each sample a he
mixing imes o 50, 100, 150, 300, and 500 ms.
3|RESULTS AND DISCUSSION
3.1 |NMR s udy o he sel -agg ega ion
p ope ies o HPC samples
The mal gela ion is a gene al p ope y o aqueous HPC
solu ions (as shown in Figu e 1) due o which, upon
hea ing, aqueous solu ions p ecipi a e and o m gels.
Gel o ma ion is a comple ely e e sible p ocess and
gelled samples lique y once again upon cooling.
[9,10]
This
cha ac e is ic depends p ima ily on molecula weigh ,
deg ee o me hyl and hyd oxyp opyl subs i u ion, con-
cen a ion, and he na u e o addi i es. A he molecula
le el, gela ion is a consequence o he o ma ion o sel -
agg ega es o HPC molecules in solu ion due o a o able
hyd ophobic in e ac ions.
[9,10]
The sel -agg ega ion abili y o he se en chosen HPC
samples was s udied a a cons an concen a ion using
wo ypes o NMR. The i s is conce ned wi h eco ding
he DOSY spec a o measu e he D alues o he mos
abundan species in solu ion. Depending on sample cha -
ac e is ics, he alue o Dob ained unde a ce ain se o
expe imen al condi ions (concen a ion and empe a u e)
e lec s ei he he monome species o an a e age o he
se e al ypes o sel -agg ega es o med in solu ion. The
a e age alues o Dob ained a e shown in Table 1.
A ep esen a i e DOSY spec um o HPC-H is shown
in Figu e 2. The D alues shown in Table 1 a e sensi i e
o M
W
. The iscosi y o hese polyme s was measu ed by
analy ical echniques such as GPC. Fu he mo e, he
in e se co ela ion be ween iscosi y and M
w
can also be
in e ed om he gi en alues.
Howe e , i he NMR alues o Da e con e ed
in o molecula weigh (M
wNMR
) using he empi ical ela-
ionship p oposed by Viel e al. o polysaccha ides
(D=8.2×10
−9
M
w
–0.49
),
[12]
he esul s in Table 1 show a
la ge disc epancy be ween M
wNMR
and M
w
, as measu ed
by GPC. M
wNMR
alues a e o e es ima ed, which s ongly
sugges s he possibili y o sel -agg ega ion unde he expe i-
men al condi ions used o eco ding NMR spec a in D
2
O.
FIGURE 1 Idealized chemical s uc u e o HPC [Colo igu e
can be iewed a wileyonlinelib a y.com]
1634 MARTIN-PASTOR AND STOYANOV
FIGURE 2 Measu emen o he di usion coe icien o HPC-H p epa ed in D
2
O om i s DOSY spec um [Colo igu e can be iewed
a wileyonlinelib a y.com]
TABLE 1 Compa ison o he analy ical and NMR da a o he se en chosen HPC samples
HPC-VH HPC-H HPC-M HPC-L HPC-SL HPC-SSL HPC-UL
M
w
(kDa)
a
2,500 1,000 700 140 100 40 20
A g. monome s 3,100 1,200 870 170 120 50 25
Viscosi y (mPa·s) 4,001–6,000 1,000–4,000 150–400 6.0–10.0 3–5.9 2.0–2.9 1.0–1.9
NMR DOSY D(×10
−10
m
2
/s) <0.01 0.02 0.05 0.12 0.18 0.34 0.48
M
wDOSY
(kDa)
b
>20,000 20,000 3,630 600 250 72 36
NMR VT-
1
H Enh_py anose 1.81 3.76 3.76 1.06 1.09 1.46 1.11
NMR VT-
1
H Enh_alipha ic 1.38 1.34 1.38 1.34 1.20 1.06 0.93
NMR alipha ic o py anose a io
c
1.29 1.13 1.24 0.99 0.85 0.81 0.79
NMR hyd a ion %H
2
O hyd
d
9% 1% 1% 81% 81% 92% 94%
NMR hyd a ion %alipha ics
e
36% 93% 94% 8% 7% 2% 0%
Abb e ia ions: HPC, hyd oxyp opyl cellulose; VT, a iable empe a u e; Wa e LOGSY,
1
H wa e ligand obse ed ia g adien spec oscopy.
a
Measu ed by GPC.
b
M
W
NMR is he molecula weigh calcula ed om he di usion coe icien D, which is measu ed om he DOSY spec um by using he
empi ical ela ionship o polysaccha ides, namely: D= 8.2 ×10
−9
M
w
–0.49
.
[11]
c
Calcula ed by he in eg a ion o he
1
H quan i a i e spec a (Figu e 3a), which is achie ed by di iding he in eg al o he alipha ic signal by
he in eg al o he py anose ing signal in he same spec um.
d
De e mined om he
1
H Wa e LOGSY spec a by s udying he poin s (h)–(n) in Figu e 4 a 50 ms. Signal in eg al co esponding o H
2
O
(hyd ) is di ided by he o al a ea o signals in he same spec um and is exp essed in %.
e
De e mined om he
1
H Wa e LOGSY spec a by s udying he poin s (h)–(n) in Figu e 4 a 50 ms. The in eg al o he alipha ic signal is
di ided by he o al signal a ea in he same spec um and is exp essed as %.
MARTIN-PASTOR AND STOYANOV 1635
To cla i y he issue o sel -agg ega ion and he size o
he species o med in he NMR samples, a iable empe -
a u e (VT)
1
H NMR quan i a i e spec a we e eco ded.
We p opose ha in he solu ion s a e, he NMR signals o
la ge species a e b oad due o hei excessi ely slow o a-
ional dynamics. Thei b oad signal leads o a educ ion
in he maximum peak in ensi y and as a esul , a pa
o o he comple e a ea o he peak may be below he
noise le el o he spec um and is pa ially o comple ely
“in isible.”This is an e ec ha has been p e iously
epo ed o se e al polysaccha ides.
[13]
Signals in he so-
called in isible egime ob iously do no con ibu e o he
expe imen al in eg al measu ed in he spec um and
hence we only obse e an appa en in eg al.
The VT echnique is use ul o s udying dynamic p o-
cesses in NMR and in his s udy, we employed i o
de ec he in isible signals o la ge species. In gene al,
empe a u e modula es he o a ional dynamics o mole-
cules in a solu ion. Upon inc easing he empe a u e, he
o a ional dynamics o la ge species a e expec ed o
inc ease and as a esul , i is possible o a pa o i s in e-
g al o shi om he in isible egime o he isible
egime causing a measu able inc ease in he (appa en )
in eg al o he signals in he spec um. The si ua ion
is simila in la ge sel -agg ega es, bu he e ec o
empe a u e on he appa en in eg al is mo e d ama ic
due o he inc ease in o a ional dynamics; his may also
a ec hei sel -agg ega ion equilib ium, causing de-
agg ega ion o a ce ain ex en . In he VT
1
H NMR spec-
a o he se en HPC samples dissol ed in D
2
O, he in e-
g al a ea o he HDO peak was measu ed along wi h he
a ea o he py anose ing p o ons and alipha ic p o ons
a 25 and 40C. The HDO peak signal was used o no -
maliza ion, as he a ea o his signal is no expec ed o
change signi ican ly in he ela i ely dilu ed samples
(1.8%) used o es ing. The case o HPC-H in Figu e 3
exempli ies he esul s o VT
1
H NMR s udies ob ained
o he se en HPC samples conside ed.
Inc easing he empe a u e om 25 o 40Ccausesa
no able inc ease in he peak in eg als co esponding o he
py anose ing ( ac o 3.76 imes in Table 1) and alipha ic
chain ( ac o 1.35 imes in Table 1) (Figu e 3). In e es ingly,
hese esul s show ha a 40C, he inc ease in he peak-
in eg al a ea was 2.8 imes mo e o he py anose ing
when compa ed o alipha ic chains; his obse a ion can be
a ibu ed o di e ences in hei lexibili y (i.e., ela i ely
as e o a ional dynamics due o in e nal mo ion) when
o ming agg ega es. The py anose ings a e aking pa
in he agg ega ion and a e mo e igid (o lexible) a he
lowe (o highe ) empe a u es. A simila phenomenon
FIGURE 3 VT
1
H spec um o HPC-H a (a) 25C, (b) 40C, and (c) supe imposi ion o he spec a a 25 and 40C wi h he same
e ical scale o he HDO peak. In (a, b), he in eg als o signals co esponding o HDO, py anose ings, and alipha ic chains a e shown;
hese signals a e no malized wi h espec o he HDO peak [Colo igu e can be iewed a wileyonlinelib a y.com]
1636 MARTIN-PASTOR AND STOYANOV

was epo ed du ing NMR analysis o he pendan chains
o ace yl g oups in pa ially ace yla ed chi osan.
[13]
The
enhancemen ac o s ob ained om he VT
1
H quan i a i e
spec a o he se en HPC samples a e lis ed in Table 1. The
end obse ed in all hese esul s was simila ; hey showed
ei he no change in he deg ee o agg ega ion in he sample
(signal enhancemen ac o o ~1) o a signi ican educ ion
in agg ega ion a highe empe a u es (signal enhancemen
ac o > 1). This unexpec ed esul con as s wi h he
epo ed e ec o gelli ica ion occu ing in o he ypes o
HPC samples a 40Cbu no a 25
C.
3.2 |NMR analysis o hyd a ion in HPC
samples
The majo cha ac e is ics o cellulose a e wa e e en ion,
adhesion, and lub ica ion. In his con ex , HPCs exhibi a
FIGURE 4 Spec a
1
H
quan i a i e ( op panel) and
Wa e LOGSY (mixing ime o
50 ms) (lowe panel) o (a, h)
HPC-UL, (b, i) HPC-SSL, (c, j)
HPC-SL, (d, k) HPC-L, (e, l)
HPC-M, ( , m) HPC-H, and (g,
n) HPC-VH dissol ed in
H
2
O. F om op o bo om,
spec a a e a anged in he
inc easing o de o M
w
[Colo
igu e can be iewed a
wileyonlinelib a y.com]
MARTIN-PASTOR AND STOYANOV 1637
good wa e -holding pe o mance and NMR can be used
o analyze hei wa e - e en ion p ope ies and dynamics
o wa e exchange. Two NMR echniques we e used in
his s udy, iz.
1
H quan i a i e o quan i y hyd a ion in
closely packed laye s and Wa e LOGSY analysis
[6]
o
in o ma ion on wa e dynamics; he la e had been used
p e iously o s udy o hyd a ion in la ge biopolyme s
such as p o eins.
[14,15]
All he samples o his s udy we e
p epa ed a a concen a ion o 1.8%; howe e , H
2
O was
used as he sol en ins ead o D
2
O.
Spec a a and b in Figu e 4 show he
1
H quan i a i e
and Wa e LOGSY spec a, espec i ely, o he se en sam-
ples s udied.
In e es ingly, all he spec a included a ela i ely b oad
peak in he egion o 5–6 ppm, which is no isible in he
1
H spec um o he analogous samples p epa ed in D
2
O
( o compa ison, see he p o on spec um in Figu es 1 and
2). The b oad peak can be a ibu ed o hyd a ion o wa e
(H
2
O
hyd
); i s p esence deno es molecules o wa e wi h a
high esidence ime in con ac wi h he HPC polyme .
Spec a a and b in Figu e 4 con ain a peak co esponding
o wa e a he expec ed posi ion (~4.7 ppm) and his signal
co esponds o he mos abundan and undi e en ia ed
bulk wa e (H
2
O
bulk
); i s in ensi y was expe imen ally
supp essed o subs an ially a enua ed o yield a dynamic
ange o de ec o he peaks.
In eg al quan i ica ion in he
1
H quan i a i e spec a
(a–g in Figu e 4) was pe o med wi h espec o he e e -
ence in eg al o py anose ing p o ons. This calcula ion
p o ides a ele an pa ame e ha is ep esen a i e o
he alipha ic o py anose a io. The ob ained esul s a e
included in Table 1 and hey exhibi a sligh and consis-
en dec ease in he o al alipha ic chain subs i u ion
acco ding o an in e se ela ionship wi h HPC M
w
. The
inc ease in he alipha ic con en can in e e e wi h he
hyd a ion p ope ies o he sample because he ex a
s e ic e ec s a e obse ed o educe he accessibili y o
wa e o he alipha ic chains e en mo e; his is in con-
as wi h he expec ed beha io o he highly hyd o-
philic pa o py anose ings.
Wa e LOGSY in ensi ies we e modula ed by he e i-
cien ans e o magne iza ion om H
2
O
bulk
o o he
ypes o p o ons in con ac wi h bulk wa e . A he
sho es mixing ime (50 ms) o Wa e LOGSY spec os-
copy (spec a h–n in Figu e 4), H
2
O
bulk
ans e s e y li -
le magne iza ion ia NOE o he signals o he py anose
ing p o ons and e en less o he alipha ic p o ons. Peak
in ensi y co esponding o H
2
O
hyd
was clea ly ampli ied
wi h espec o o he peaks in he same spec um and
e en he same peak in he co esponding
1
H quan i a i e
spec a (a–g in Figu e 4). The la e obse a ion implies
a ema kable ans e o magne iza ion om exci ed
H
2
O
bulk
in Wa e LOGSY spec a o H
2
O
hyd
du ing he
sho mixing ime (50 ms) used o eco d he spec a. In
be ween he wo possible NOE and chemical exchange
mechanisms ha con ibu e o signal in ensi y in Wa e -
LOGSY spec a, he la e esul s in conside ably as e
kine ics and con ibu es o a la ge ex en o he in ensi y
o he H
2
O
hyd
peak. Scheme 1 illus a es he mechanism
o chemical exchange equilib ium ha con e s he Wa e -
LOGSY exci ed signal o H
2
O
bulk
wa e in o ha o
H
2
O
hyd
. This mechanism ollows wo kine ic cons an s k
1
and k
−1
ha ollow he ela ionship k
1
>> k
−1
; u he -
mo e, k
1
>50ms.
The e olu ion o Wa e LOGSY signal in ensi ies a a
se ies o mixing imes is ep esen ed in Figu e 5 o he
se en chosen HPC samples.
The peak beha io obse ed in Figu e 5 was
ex emely simila o all he se en samples. The in ensi y
o he H
2
O
hyd
peak decayed wi h an inc ease in mixing
ime, which is in ag eemen wi h he exchange mecha-
nism p oposed in Scheme 1. In con as , he in ensi ies o
SCHEME 1 Wa e -HPC equilib ium in solu ion and Wa e LOGSY p inciple. The magne iza ion o he bulk wa e molecules H
2
O
A_bulk
is selec i ely aken ou o equilib ium (in e ed) by Wa e LOGSY. Following equilib ium, H
2
O
A_bulk
molecules a e in oduced o he
hyd a ion laye o HPC, and p e ious wa e molecules H
2
O
B_hyd
a e ejec ed. Obse able in e molecula NOEs a e gene a ed when he
esidence ime o H
2
O
A_hyd
in con ac wi h HPC is su icien ly long, which equi es he kine ic condi ion k
1
>k
−1
[Colo igu e can be
iewed a wileyonlinelib a y.com]
1638 MARTIN-PASTOR AND STOYANOV
he py anose ing p o ons and alipha ic chains inc ease
wi h an inc ease in mixing ime, which is consis en wi h
he NOE mechanism media ed by he p oximi y o hese
p o ons o he wo ypes o wa e , H
2
O
bulk
and mo e
impo an ly, H
2
O
hyd
.
Ha ing iden i ied he e ec s ha occu wi h espec
o peak in ensi y in Wa e LOGSY spec a, peak in eg als
(h)–(n) in Figu e 4) we e quan i ied a he lowes mixing
ime o 50 ms. These calcula ions yielded wo indepen-
den pa ame e s ela ed o hyd a ion. One pa ame e is
FIGURE 5 Wa e LOGSY
in ensi y build-up plo o signals
measu ed o HPC samples
p epa ed in H
2
O. (a) HPC-UL,
(b) HPC-SSL, (c) HPC-SL,
(d) HPC-L, (e) HPC-M, ( ) HPC-H,
and (g) HPC-VH. The a ea
co esponding o py anose ing
p o ons was in eg a ed in ou
di e en egions as indica ed by
pa en heses (in ppm) [Colo igu e
can be iewed a
wileyonlinelib a y.com]
MARTIN-PASTOR AND STOYANOV 1639
he ela i e a ea o peak H
2
O
hyd
wi h espec o he o al
signal a ea in he same spec um (%H
2
O
hyd
). The highes
alues o %H
2
O
hyd
(see Table 1) we e ob ained o HPC
samples wi h a M
w
o 140 kDa o less (81–94%). Fo hose
samples wi h highe M
w
, he alues o %H
2
O
hyd
we e no a-
bly lowe (1–9%). The second pa ame e is he ela i e a ea
o he alipha ic signals wi h espec o he o al signal a ea
in he same spec um (%alipha ics). I can be seen in
Table 1 ha he end o %alipha ics is he opposi e o ha
o %H
2
O
hyd
; a e y small %alipha ics alue was obse ed
a M
w
alues 140 kDa o lowe (0–8%) while no ably high
alues we e ob ained a highe M
w
alues (36–94%).
The ends obse ed o %H
2
O
hyd
and %alipha ics
can be explained as ollows; H
2
O
hyd
e lec s he amoun
o wa e ha he polyme is able o in oduce in he sol-
a ion laye o he HPC s uc u e. I mus be disposed
owa d he ex e nal side o he polyme o ensu e ha
H
2
O
hyd
can exchange wi h ex e nal bulk wa e . As he
leng h o he HPC polyme chain inc eases, he con ibu-
ion o %H
2
O
hyd
becomes less ele an because o he
pa s o he HPC s uc u e con ibu e mo e o wa e
e en ion; in pa icula , he e ec o alipha ic esidues,
exp essed by %alipha ics in Table 1, becomes he mos
p ominen ac o a M
w
> 140 kDa. This obse a ion can
be explained by conside ing ha he hyd ophobici y o
alipha ic esidues o HPC can lead o he o ma ion o
s able esicles on he in e io side o he HPC polyme .
These esicles can e ain a subs an ial amoun o wa e
wi hou exchanging wi h he bulk wa e loca ed on he
ex e io side.
The cu es in Figu e 5 o he se en chosen HPC sam-
ples co esponding o he py anose p o ons a e obse ed
o be abo e he cu es o he alipha ic p o ons, which
indica e ha he o me a e mo e exposed o he wa e .
4|CONCLUSIONS
In his wo k, we applied liquid NMR me hods o s udy
HPC samples in equilib ium-swollen gels p epa ed a a
cons an concen a ion in he low- iscosi y egime.
[11]
The esul s ob ained he ein a e consis en and comple-
men he p e ious s udies on HPC, which shows ha sol-
ubili y is enhanced o sho -chain g ades.
[1–3]
In his
wo k, we ind ha he wa e e en ion o HPC, and he e-
o e i s d ug solubili y, is modula ed p ima ily by he
polyme chain leng h (M
w
) and he alipha ic chain con-
en . E iden ly, HPCs wi h he lowes molecula weigh s
(UL and SSL) exhibi an associa ion wi h he ac ha
hey comp ise he lowes amoun o wa e molecules
a ound hei polyme chains; his is p obably because
hey do no allow signi ican esicle o ma ion and conse-
quen ly, exhibi subs an ially low agglome a ion. This
phenomenon could “ acili a e” he in e ac ion be ween
HPC-UL and SSL, hus p e en ing he p ecipi a ion o
small hyd ophobic molecules in aqueous solu ions.
P e ious me hods o s udying HPC hyd a ion based
on NMR longi udinal (T
1
) and ans e se (T
2
) pa ame e s
ha e analyzed he bulk wa e peak, and di e en ypes
o wa e o de ing in hyd ogels o HPC we e epo ed.
These s udies ound wo ypes o wa e molecules, bulk-like
wa e ( eezable wa e a 0C) and wa e weakly in e ac ing
wi h mac omolecules (non eezable wa e a 0C).
[16]
The
me hod p oposed in his wo k, which elies on VT
1
H-
quan i a i e and Wa e LOGSY spec a, is consis en wi h
hose indings and has pe mi ed he de ec ion o wo inde-
penden esonances o he bulk wa e and hyd a ed wa e
(Figu e 4). These indings e lec he di e ences in he mag-
ne ic suscep ibili y be ween he en i onmen s o he wo
hyd a ion modes, which a e in slow exchange on he chem-
ical shi imescale unde ou expe imen al condi ions.
S udies based on MRI and NMR elaxa ion ha e
de ec ed a g adien o wa e mobili y ac oss he gel laye
and di e en deg ees o polyme hyd a ion a di e en
dep hs in he gel.
[17]
Howe e , he Wa e LOGSY used in
his wo k is sensi i e only o di ec molecula con ac
be ween wa e and HPC. This sensi i i y implies ha he
peak in ensi y in he Wa e LOGSY spec um speci ically
e lec s he ac ion o bulk wa e wi h su icien accessi-
bili y o pene a e in o he i s hyd a ion shell o HPC
(Scheme 1), which has ob ious implica ions o he cha -
ac e iza ion o HPCs as d ug-deli e y sys ems.
Solu ion NMR is known o ha e limi a ions wi h
espec o he maximum size o he HPC molecule o
agg ega e ha can be s udied; in his sense, ou esul s
a e skewed owa d hose pa s along he HPC chain ha
exhibi a su icien molecula mobili y, which a e exac ly
he pa s ha a e mo e exposed o he su ounding wa e .
Fu u e wo ks o employing solid NMR can p o ide he
complemen a y in o ma ion ega ding he mos igid
pa s o HPC as a unc ion o wa e con en .
ORCID
Edmon S oyano h ps://o cid.o g/0000-0002-1571-
9259
REFERENCES
[1] A. I o, C. Konne h, J. Schmid , W. Peuke , Eu . J. Pha m.
Biopha m. 2016,98, 98.
[2] C. Konne h, F. Flach, S. B ei ung-Faes, C. Damm, J. Schmid ,
A. Kwade, W. Peuke , Powde Technol. 2016,294, 71.
[3] D. E. Zece ic, R. Meie , R. Daniels, K. G. Wagne , Eu .
J. Pha m. Biopha m. 2014,87, 264.
[4] K. Kimu a, T. Shigemu a, M. Kubo, Y. Ma u, Mak omol.
Chem. Mac omol. Chem. Phys. 1985,186, 61.
[5] D. Desai, F. Rinaldi, S. Ko ha i, S. Pa uchu i, D. Li, M. Lai,
S. Fung, D. Bo h, In . J. Pha m. 2006,308, 40.
1640 MARTIN-PASTOR AND STOYANOV