Synthesis, structure, chemical and bioactivity behavior of eight chromium(III) picolinate derivatives Cr(R-pic)3

Article abstract of DOI:10.1016/j.ica.2017.05.041

The worldwide use of chromium(III) picolinate Cr(pic)₃ as nutrition additives has aroused more and more controversies. To reevaluate the?safety and validity of Cr(pic)₃, seven new derivatives Cr(R-pic)₃ (pic?=?picolinic acid, R?=?H (1), 5-Br (2), 5-CF₃ (3), 4-Cl (4), 5-COOH (5), 3-CH₃ (6), 5-OH (7), 3-OH (8)) were synthesized and characterized by X-ray crystal diffraction, ESI-MS, IR and elemental analysis. It was found that different substituent group affected physicochemical activities of the complex such as the Fenton-like reaction and oxidation reaction. Especially, –OH group derivatives lose their hydroxyl radical-generation and Cr(VI)-generation abilities comparing with halogen group in tube experiment. Even so, these differences in chemistry properties may be ignored in live cells and animal tests: no obvious cellular?damage (MTT assay) and tissue injury (acute toxicity study) were observed for both Cr(pic)₃ and its derivatives. In addition, hypoglycemic activity study indicated that these Cr(III) complexes have no significant influence than CrCl₃ salt on the blood glucose, serum insulin, total cholesterol, triglyceride, high density lipoprotein and low density lipoprotein of diabetic mice through two months’ study. Therefore, these substituent group is unable to improve the biological activities of Cr(pic)₃ obviously and the validity of Cr(pic)₃ used as a nutrition additives is doubted.

Full text of DOI:10.1016/j.ica.2017.05.041

Inorganica Chimica Acta 466 (2017) 151–159
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Synthesis, structure, chemical and bioactivity behavior of eight
chromium(III) picolinate derivatives Cr(R-pic)3
⇑
⇑
Jie Chai, Yanfei Liu, Jinglong Dong, Bin Liu , Binsheng Yang
Institute of Molecular Science, Key Laboratory of Chemical Biology of Molecular Engineering of Education Ministry, Shanxi University, Taiyuan 030006, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 April 2017
Received in revised form 19 May 2017
Accepted 22 May 2017
Available online 25 May 2017
The worldwide use of chromium(III) picolinate Cr(pic)3 as nutrition additives has aroused more and more
controversies. To reevaluate the safety and validity of Cr(pic)3, seven new derivatives Cr(R-pic)3
(pic = picolinic acid, R = H (1), 5-Br (2), 5-CF3 (3), 4-Cl (4), 5-COOH (5), 3-CH3 (6), 5-OH (7), 3-OH (8)) were
synthesized and characterized by X-ray crystal diffraction, ESI-MS, IR and elemental analysis. It was
found that different substituent group affected physicochemical activities of the complex such as the
Fenton-like reaction and oxidation reaction. Especially, –OH group derivatives lose their hydroxyl radi-
cal-generation and Cr(VI)-generation abilities comparing with halogen group in tube experiment. Even
so, these differences in chemistry properties may be ignored in live cells and animal tests: no obvious cel-
lular damage (MTT assay) and tissue injury (acute toxicity study) were observed for both Cr(pic)3 and its
derivatives. In addition, hypoglycemic activity study indicated that these Cr(III) complexes have no sig-
nificant influence than CrCl3 salt on the blood glucose, serum insulin, total cholesterol, triglyceride, high
density lipoprotein and low density lipoprotein of diabetic mice through two months’ study. Therefore,
these substituent group is unable to improve the biological activities of Cr(pic)3 obviously and the validity
of Cr(pic)3 used as a nutrition additives is doubted.
Keywords:
Chromium(III) picolinate
Fenton-like reaction
Oxidation reaction
Cell damage
Hypoglycemic activity
Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction
exhibits well effect [9,10]. To improve the biological activity and
lower the toxicity of Cr(III) complexes, varieties of new complexes
Cr(III) was proposed to be an essential element for its efficiency
in carbohydrate and lipid metabolism in the 1950s [1], and it has
been suggested to be an artificial second messenger to amplify
insulin signal [2]. As a man-made complex, Cr(pic)3 has become
one of the best-selling food additives [3], and a plenty of investiga-
tions about it have been made to better understand the role of Cr
(III) in anti-diabetes. Till now, the validity and safety of Cr(pic)3
remains disputable [4] for two reasons: (1) Cr(pic)3 cannot be bio-
logically active form of Cr(III) for the complex possessing no inher-
ent biological activity itself [5], and (2) previous studies have
revealed that Cr(pic)3 can generate hydroxyl radical (ÅOH) in the
presence of H2O2/vitamin C to cleavage DNA [6]. Thus the status
of Cr(III) as an essential element has recently been challenged,
and the essentiality of Cr(III) is a matter of current debate [7].
Despite no acute toxicity was observed in vivo through two
years’ animal experiments [8], Cr(pic)3 was supposed to be unsafe.
There is increasing evidence that the risk of Cr(pic)3 derives from
the ligands for picolinate is mildly clastogenic by itself [6]. Hence,
picolinate is always used as ligands for anticancer complexes and
such as [Cr(D-Phe)3] [11], NBC [12], CDNC [13,14], LMWCr [15], [Cr
(met)+3] [16,17] and CrSA [18] have been synthesized. In addition,
the toxicity and hypoglycemic activity of these Cr(III) complexes
are evaluated by animal study. We previously reported a direct
relation between ÅOH generation and the physicochemical proper-
ties of Cr(pic)3 derivatives by synthesis and research of three new
Cr(pic)3 derivatives, Cr(6-CH3-pic)3, Cr(3-NH2-pic)3, and [Cr(6NH2-
pic)2(H2O)2]ÁNO3 [19]. However, systematic studies on the safety
and activity of Cr(pic)3 and its derivatives are still needed.
The present study provides seven new chromium(III) picolinate
derivatives Cr(R-pic)3 (pic = picolinic acid, R = H (1), 5-Br (2), 5-CF3
(3), 4-Cl (4), 5-COOH (5), 3-CH3 (6), 5-OH (7), 3-OH (8)), physico-
chemical properties and biological activity of these complexes.
Firstly, ÅOH and Cr(VI) generation of Cr(pic)3 was measured by Fen-
ton-like reaction [20] and H2O2 oxidation respectively in vitro in
neutral aqueous media (pH 7.4) [21]. Then, cell damage, acute tox-
icity and hypoglycemic activity of Cr(pic)3 derivatives were carried
out. Herein, we are trying to improve the safety and hypoglycemic
activity of Cr(pic)3 by the substituent effect, and the results of our
investigations will try to reevaluate the use of chromium picolinate
complexes as food additives.
⇑
Corresponding authors.
E-mail addresses: liubin@sxu.edu.cn (B. Liu), yangbs@sxu.edu.cn (B. Yang).
http://dx.doi.org/10.1016/j.ica.2017.05.041
0020-1693/Ó 2017 Elsevier B.V. All rights reserved.

152
J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
allowing the filtrate of the mixture stand for 1 week at room tem-
2. Experimental
2.1. Materials and chemicals
perature. These complexes were characterized by IR, ESI-MS, ele-
mental analysis, and/or X-ray single-crystal diffraction (XRD)
(Figs. S1–S4). The detailed synthesis process and structure charac-
terization of complexes 1–8 by IR, ESI-MS and elemental analysis
are shown in the Supporting information (Figs. S5–S12 and
Figs. S21–S25 for IR and ESI-MS respectively).
Cr(NO3)3Á9H2O, CrCl3Á6H2O, H2O2, ascorbic acid, 2-thiobarbi-
turic acid, phosphate and all the R-pic ligands including pyridine-
2-carboxylic acid (pic, L1), 5-bromine-2-carboxylic acid (5-Br-pic,
L2), 5-(trifluoromethyl)pyridine-2-carboxylic acid (5-CF3-pic, L3),
4-chloropyridine-2-carboxylic acid (4-Cl-pic, L4), 2,5-pyridine-2-
carboxylic acid (5-COOH-pic, L5), 3-methylpyridine-2-carboxylic
acid (3-CH3-pic, L6), 5-hydroxypicolinic acid (5-OH-pic, L7), 3-
hydroxy-2-pyridinecarboxylic acid (3-OH-pic, L8) were obtained
from Aladdin Chemistry Co. Ltd. (Shanghai, China). Dulbecco min-
imum essential media (DMEM), RPMI medium 1640, trypsin diges-
tive juices, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide (MTT), and fetal bovine serum were from Solarbio Science
& Technology Co. LTD (Beijing, China). Unless other indicated, all
the reagents were used without purification.
Crystallographic data of Cr(III) complexes were collected on a
D8 Venture X-ray diffractometer (Bruker, Germany). ESI-MS spec-
tra were recorded on an Agilent 6520 Accurate-Mass QTOF LC/
MS mass spectrometer in DMSO. UV–visible (UV–vis) spectra and
elemental analyses were estimated on a Varian 50 Bio spectropho-
tometer and a Vario EL III analyzer respectively. IR spectra were
measured with a Bruker TENSOR 21 FT-IR spectrophotometer.
Water-jacketed CO2 cell incubator was from Shanghai Lishen Sci-
entific Instruments Co., Ltd. The Animal center of First Hospital of
Shanxi Medical University (Shanxi Province, China) approved the
animal experimental protocols with the qualified number SCXK
(JIN) 2015–0001.
2.3. X-ray diffraction
The XRD measurements for complexes 1, 2, 4, and 6 were car-
ried out with a CCD detector equipped with graphite-monochrome
Mo Ka radiation. The structures of all complexes mentioned above
were obtained by direct methods with SHELXL-97 and refined by
full-matrix least-squares techniques on F2 with SHELXL-97 [23].
The ORTEP diagram of complexes 1, 2, 4 and 6 were demonstrated
in Fig. 1, and the crystallographic data, structure refinements, bond
lengths and angles were listed in Table 1 and Table 2 (CCDC num-
bers for complexes 2, 4, and 6: 1500350, 1500280 and 1500281).
2.4. Fenton-like reaction
Hydrogen peroxide (1.0 Â 10À4 M) was added to the mix solu-
tion of medium (PBS, 1640 or HSA)/Fe(EDTA)/complexes 1–8
(1.0 Â 10À4 M, PBS), D-2-Deoxyribose (4.0 Â 10À3 M) and ascorbic
acid (1.0 Â 10À4 M) in PBS/RMPI medium 1640/HSA (1 Â 10À4 M,
PBS). After 3 h of water bath (37 °C), 2-thiobarbituric acid (2.8%
w/V, 300 mL) and trichloroacetic acid (1% w/V, 5 mL) were added
to the reaction mixture (2 mL) consecutively. The final solution
turned pink from colorless with a maximum absorbance at
532 nm after 30 min of water bath (90 °C) [20]. In this experiment,
Fe(EDTA) act as control group and medium only as blank.
2.2. Preparation of Cr(R-pic)3
Cr(pic)3Á(H2O) (1) were prepared according to the literature
[22]. Cr(R-pic)3 complexes 2–8 were obtained by the similar meth-
ods (Scheme 1). A solution of Cr(NO3)3Á9H2O (0.20 g, 0.50 mmol)
was added to the mixed solution of R-pic ligands (1.5 mmol) and
NaOH (1.5 mmol, 20 ml) dropwise and refluxed for 1 h. Pink crys-
tals suitable for X-ray single-crystal diffraction were obtained after
2.5. Oxidation reaction
With the addition of complexes 1–8 (1 Â 10À4 M, PBS, pH 7.4) to
the solution (PBS/RPMI medium 1640/HSA) of H2O2 (1 Â 10À4 M,
PBS buffer: pH 7.4), the tubes were put into water bath (37 °C)
for 3 h. With the addition of 1, 5-diphenylcarbazide (1.0 mL,
Cl
F3C
Br
O
O
O
O
N
Cr
O
N
Cr
O
N
Cr
O
N
Cr
O
O O
N
O
O
N
O O
N
O
N
O
N
O
N
O
O
N
O
N
CF3
Br
Cl
Cl
CF3
O
O
O
Br
O
C18H9Cl3CrN3O6
C18H12CrN3O6
C18H9Br3CrN3O6
C21H9CrF9N3O6
1
2
3
4
CH3
O
OH
O
O
N
HO
O
HO
N
O
O
N
O
N
O O
O
N
O
O
O
O
Cr
O
O O
O
O
H3C
Cr
N
Cr
N
HO
Cr
OH
N
N
N
N
N
OH
O
O
O
O
O
CH3
O
OH
O
OH
HO
O
C18H12CrN3O9
C18H12CrN3O9
C21H12CrN3O12
C21H18CrN3O6
5
6
7
8
Scheme 1. The structure of Cr(R-pic)3 complexes 1–8.

J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
153
Fig. 1. Crystal structures of complexes 1(a), 2(b), 4(c) and 6(d).
Table 1
Crystal data and structure refinement for complexes 1 (a), 2 (b), 4 (c) and 6 (d).
Complex 1
Complex 2
Complex 4
Complex 6
Empirical formula
Formula weight
CCDC number
Crystal system
Space group
a/Å
C18H14CrN3O7
436.32
C18H9Br3CrN3O6
655.01
1500350
Monoclinic
P2/n
9.772
17.776
13.084
C18H9Cl3CrN3O6
521.63
1500280
Tetragonal
I4 cd
29.632
29.632
9.316
C21H18CrN3O6
460.38
1500281
Monoclinic
C2/c
31.764(3)
8.5367(8)
14.9007(15)
90
Monoclinic
C2/c
30.196
8.5360
13.9321
90
b/Å
c/Å
a
(°)
90
90
b(°)
95.0160
90
3577.3
110.43
90
2129.8
4
90
90
8179.8
16
96.592(3)
90
4013.8(7)
8
c
(°)
V/Å3
Z
8
d g/cm3
1.620
2.043
1.694
1.524
F(000)
1784
1260
4176
1896
h
2.48–25.02
À35/34, À10/10, 6/12
8747/3158/0.0635
3158/262
1.014
2.83–28.26
À8/13, À23/23, À17/17
20070/5251/0.0553
5251/280
3.07–28.32
À35/39, À39/37, À12/12
41200/5091/0.1086
5091/280
2.86–28.32
h, k, l
Tot.Uniq.Data Rint
À42/42, À11/7, À19/19
19709/4973/0.0334
4973/283
N
reflections, Nparameter
GOF on F2
1.010
1.029
1.041
R1[I > 2s (I)],
xR2
0.0495, 0.1025
0.0451, 0.0875
0.0523, 0.0739
0.0389, 0.0980
0.01 g/mL) and concentrated sulfuric acid (20 mL) to the mixture
consecutively, the sample turned to purple from colorless, and Cr
(VI) was detected spectrophotometrically at 548 nm. The linear
responses were 0.0420–0.0015 L/mg for Cr(VI) [24,25].
complexes 1–8 (20 mL, 1 Â 10À3 M, DMSO/buffer, 1:9 V/V). Subse-
quently, MTT (20 mL, 5 mg/mL, and NaCl buffer) were added and
the plates were incubated at 37 °C for 4 h to form a purple
formazan. Finally, the medium with MTT was removed from the
wells and 200 mL of DMSO was transferred into each well to dis-
solve the purple formazan. The absorbance at 490 nm was mea-
sured with a micro plate reader and absorbance of blank control
wells were taken as 100% of viability. Results are indicated as the
percentage of viable cells with respect to untreated controls.
Values are the mean SD of three separate experiments carried
2.6. MTT assay
Cell lines at an initial density of 4.5 Â 103 cells/well in 96-well
plates containing DMEM (180 mL) were cultured for 14 h at first.
Followed by another culture for 24 h with the presence of

154
J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
Table 2
Selected bond lengths (Å) and angles (°) of complexes 1, 2, 4 and 6.
Complex 1
Complex 2
Complex 4
Complex 6
Bond length/ Å
Cr1ÀN1
2.048(3)
2.056(3)
2.057(3)
1.954(3)
1.948(3)
1.942(3)
2.084(3)
2.057(3)
2.047(3)
1.933(3)
1.948(3)
1.951(3)
2.058(3)
2.065(3)
2.042(4)
1.953(3)
1.947(3)
1.936(3)
2.0557(2)
2.0594(2)
2.0433(2)
1.9505(1)
1.9368(2)
1.9329(1)
92.42(6)
169.84(7)
95.72(6)
79.92(6)
94.32(6)
93.38(6)
95.45(7)
79.88(6)
91.97(7)
92.31(6)
171.74(6)
80.36(6)
172.48(6)
93.18(6)
92.93(6)
Cr1-N2
Cr1-N3
Cr1-O1
Cr1-O3
Cr1-O5
Bond angles (°)
N1-Cr1-N2
N1-Cr1- N3
N2-Cr1-N3
O1-Cr1-N1
O1-Cr1-N2
O1-Cr1-N3
O3-Cr1-N1
O3-Cr1-N2
O3-Cr1-N3
O5-Cr1-N1
O5-Cr1- N2
O5-Cr1-N3
O3-Cr1-O1
O5-Cr1-O1
O5-Cr1-O3
168.61(12)
92.22(12)
97.22(12)
80.30(12)
92.53(12)
94.70(12)
90.49(11)
81.17(12)
170.89(12)
99.13(12)
88.69(12)
80.95(12)
94.33(11)
175.60(12)
90.03(12)
162.69(14)
98.89(13)
95.91(12)
81.10(12)
85.79(12)
170.50(13)
90.64(12)
80.58(12)
89.29(12)
94.47(13)
96.90(13)
80.00(12)
100.20(13)
90.53(13)
168.72(12)
93.27(13)
167.57(14)
95.85(13)
81.36(14)
94.44(13)
89.52(13)
94.06(14)
80.64(13)
95.74(13)
90.71(13)
171.01(14)
81.49(13)
173.12(13)
94.13(13)
91.04(14)
out in sextuplicate. DMSO is used as a control comparison, the final
concentration of DMSO in the experimental group and DMSO
group is limited at 1%.
Biomedicals (Costa Mesa, Calif.), respectively. As expected, FBG
levels increase approximately 1-fold following treatment with
STZ vs. non-STZ treated mice, and FBG levels ꢀ11.1 mmol/L repre-
sents the mice are hyperglycemic.
Fifty mice were randomly divided into one normal group (nor-
mal mice) and four diabetic groups (STZ-mice) with 10 mice each
group. All the mice were free access to standard solid diet and
water during the experimental process. The drug was adminis-
trated at the doses of 1000 mg Cr/kg body mass orally once a day
for 8 weeks, and the body mass, fasting blood glucose (FBG), fasting
serum insulin (FINS), total cholesterol (TC), triglyceride (TG), high
density lipoprotein (HDL) and high density lipoprotein (LDL) levels
of the mice was measured every 4 week thereafter. Blood sample of
all mice were obtained through orbital venous plexus [26].
In the end, all the samples were analyzed by Beckman AU5800
automatic biochemical analyzer features in the First Hospital of
Shanxi Medical University (Shanxi Province, China). The data were
analyzed using statistical software SPSS version 16.0 (SPSS Inc.,
Chicago, USA) and given as mean SD (standard deviation). Exper-
imental results were considered to be specific significant different
when p ꢁ 0.05.
2.7. Acute toxicity study
All the animal experimental protocols were approved by the
Animal centre of First Hospital of Shanxi Medical University
(Shanxi Province, China) with the qualified number SCXK (JIN)
2015-0001. Healthy C57 mice at the mass of 20–30 g with equal
sex were divided into five groups in this study. DMSO (1%) was
used as control, other groups were treated with CrCl3, 1, and 8,
respectively. On account of insolubility of the complexes in redis-
tilled water and other buffer solution, 1% DMSO was employed as
solvent in acute toxicity study. After fasted for 16 h, all mice were
administered orally a single dose 1% DMSO, CrCl3, 1 and 8 (10.0 g/
kg BW). Observe sostenuto the behaviors of the mice for 1 h, and
within 24 h intermittently for every 4 h. Then intragastric adminis-
tration for 6 days’ feeding with the same dosage as administered
orally. During the process, all animals are free access to water
and food. The toxic symptom and death of all mice need further
observation for 1 week [17], and the body weight at 0, 3rd and
7th day should be recorded. At the final stage of the experiment,
all the mice were killed by carbon dioxide and dissected after suf-
fering from starvation for 12 h. 3–5 mm sections of liver, kidney
and pancreas of the rats stained by hematoxylin and eosin (H&E)
were examined macroscopically. The morphological observation
of the internal organs was given visually for signs of toxicity [17].
3. Results and discussions
3.1. Description of the structure
Complexes [Cr(pic)3]Á(H2O) (1), Cr(5-Br-pic)3 (2), Cr(5-CF3-pic)3
(3), Cr(4-Cl-pic)3 (4), Cr(5-COOH-pic)3 (5), Cr(3-CH3-pic)3 (6), Cr(5-
OH-pic)3 (7) and Cr(3-OH-pic)3 (8) were synthesized and charac-
terized by infrared spectroscopy, ESI-MS and elemental analysis.
The infrared spectra of the complexes and corresponding ligands
are shown in Figs. S5–S12. The FTIR analysis in Table S1 can well
depict the coordination between L1-L8 and Cr3+. For example,
stretching vibration of carboxylate for L2 has been shifted from
1699 to 1677 cmÀ1 after coordination; meanwhile, stretching
vibration of –OH in L2 at 3398 cmÀ1 disappeared in the complex 2.
The structure of [Cr(pic)3]Á(H2O) is a mononuclear meridional
isomer, and it is almost the same with the literature reported pre-
viously [22]. Plenty of O-HÁ Á ÁO hydrogen bonds was found in the
Crystal packing diagram (Fig. S1). The crystal structure of com-
plexes 2, 4 and 6 are similar with 1. Cr(5-Br-pic)3 (2) crystallizes
in monoclinic P2/n space group (Table 1). Central metal Cr(III) ion
is chelated by three 5-Br-pic ligands with meridional configuration
2.8. Hypoglycemic activity study
After 4 weeks feeding with control (normal group) or high-fat
(diabetic group) diets: ordinary feed 51.5%, soybean oil 10%,
sucrose 20%, lard 15% and cholesterol 3.5%, mice were injected
with citric buffer (normal group) or streptozotocin (STZ in citric
buffer, pH = 4.4, 50 mg/kg body mass) through tail vein. After con-
tinuation of 4 weeks’ control or high fat diets, blood samples were
drawn and fasting glucose levels (FBG) were determined to confirm
induction of diabetes. All mice were starved for 16 h prior treat-
ment, and heparin (0.5 mg/ mL) and NaF (10 mg/mL) were added
to the blood sample immediately after the blood was collected.
The diagnostic kits, double antibody and antibody coated kits used
in this experiment were from Sigma (St. Louis, Mo.) and ICN

J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
155
and assembles in a distorted octahedron (Fig. 1b). Bond angles of
N1-Cr1-N2, O3-Cr1-O5, O1-Cr1-N3 are 162.69, 168.72, 170.50°
and the Cr–O/Cr-N distances range from 1.933(3) to 1.951(3) Å
and 2.047(3) to 2.084(3) Å (Table 2), respectively. Although no
H2O molecular were found from the crystal packing structure, effec-
literature [30]. This is because the halogen atom probably occupies
the sites that are more easily attacked by ÅOH [31,32]. Whatever, all
the derivatives generate less ÅOH comparing to Cr(pic)3 and Fe
(EDTA).
Apart from PBS medium, Fenton-like reactions were also carried
out in HSA (green) and RPMI medium 1640 (blue). As a result in
tive interlayer interactions such as CAHÁ Á Á
p interactions was
Å
formed (Fig. S2). Cr(4-Cl-pic)3 (4) is a mononuclear Cr(III) complex
crystallizes in tetragonal I4 cd space group with meridional config-
uration (Fig. 1c). Cr–O bond lengths ranges from 1.936(3) to 1.953
(3) Å and the bond lengths of Cr–N bonds range from 2.042(4) to
2.065(3) Å. In addition, structure of Cr(3-CH3-pic)3 (6) is shown in
Fig. 1d, and it is a meridional isomer as complexes 1, 2 and 4. The
Cr-N distance ranges from 2.0433 (2) to 2.0594(2) while the Cr-O
bond lengths range from 1.9329(1) to 1.9505(2) Å. This is in accor-
dance with the structure of Cr(pic)3 reported in 1992 [22]. Weaker
C-HÁ Á ÁO interactions exist in intermolecular interactions in 4 and 6
(Figs. S3, S4).
Fig. 2b, complex 8 generate less OH than Cr(pic)3 no matter in
PBS, 1640 or HSA medium. Interestingly, ÅOH-generation by all
the samples in RPMI medium 1640 and HSA is far lower than that
Å
in PBS medium. The probably reason is that OH will be captured
and consumed by the bio-reductants or proteins in 1640 or HSA
medium. This process is complex and unable to control. To sum
Å
up, Cr(III) complexes can generate OH with the existence of oxi-
Å
dant and reducing agent by Fenton-like reaction, and OH will be
consumed by bio-reductants in the medium. Therefore, the dam-
Å
age of OH to cells and tissues can be ignored.
The average Cr-N length is 2.063, 2.055, 2.053 Å, and this is
semblable with other similar complexes (2.097 [19], 2.053 [22])
The biggest bond angles of N-Cr-N(#) is 162.69, 167.57 and
169.84 ° for 2, 4 and 6 respectively, and this is alike with other
CrN3O3 complexes (166.75 [27], 174.98 [28] and 168.1° [29]).
Overall, there is no significant difference in the bond lengths and
bond angles for these complexes although with different sub-
stituent group.
Single crystals of complexes 3, 5, 7 and 8 could not be obtained,
and the molecular structures of these complexes are not definite.
We hypothesis that the structures of complexes 3, 5, 7 and 8 are
also meridional isomers based on the results of UV–vis, ESI-MS
and elemental analysis and they were obtained at the same condi-
tions with complexes 1, 2, 4 and 6. Besides, the ESI-MS of the com-
plexes in Figs. S21–S25 of complexes 2, 3, 5, 7 and 8 indentify the
formation of Cr complex, respectively.
3.3. Oxidation reaction
As is known to us all, Cr(VI) is mutagenic agent and exerts
potential hazard to organism [33]. To explore the oxidation poten-
tial of the Cr(III) complexes, oxidation reaction by hydrogen perox-
ide in PBS, RPMI medium 1640 and HSA were monitored by UV–vis
spectra. The results of oxidation reaction were given in Fig. 3, and
the peak appeared at 548 nm reflects the generation of Cr(VI) in the
solution. As seen in Fig. 3b, complexes 2–4 generate more Cr(VI)
than other complexes, and we attribute this to the electronegative
group on the ligands (bearing –Br, –CF3, –Cl). Besides, Cr(VI)-gener-
ation of 7 and 8 (bearing -OH group) is least and insusceptible to
Å
medium, which is in accordance with the OH-generation in Fen-
ton-like reaction. Besides, most of complexes generate much more
Cr(VI) in 1640 or HSA medium than that in PBS, and it is because
that the HSA or 1640 medium may act as a competitor to make
the complexes more easy to decompose and be oxidized. Besides,
the reducing agents in HSA and 1640 are able to consume H2O2,
and the unequal capture ability of H2O2 by bio-reductants results
in different Cr(VI)-generation in the medium, and this could
3.2. Fenton-like reaction
It is reported that free radical ÅOH generated by Cr(pic)3 will do
harm to chromosome. In order to evaluate the safety of complexes
Å
explain why so big errors in Fig. 3b. As a result of OH-generation
Å
and Cr(VI)-generation reaction, complexes contain –OH group
behaves less in vitro toxicity potential than that of halogen group.
In addition, the direct oxidation to Cr(VI) by H2O2 in cell cultural
medium should be paid more attention than Fenton-like reaction
for Cr(pic)3 and halogen group complexes.
1–8 in vitro, OH-generation of these complexes were evaluated
firstly by traditional Fenton-like reaction in PBS medium, and the
Å
absorbance at 532 nm of the samples indirectly reflects the OH-
Å
generation [20]. As shown in Fig. 2a, OH-generation by different
complexes is various in identical conditions. In comparison with
Å
1–6, 7 and 8 generate much less OH, which may be due to the
–OH group on ligands. Perhaps phenolic hydroxyl –OH group is
3.4. MTT assay
Å
able to capture OH, which is confirmed by Fenton-like reaction
Å
of ligands alone (Fig. S26). Besides, 2–4 generate much more OH
Cytotoxicity of complexes 1–8 was tested by MTT (3-(4,
than 6, which may result from halogen group according to previous
5-dimethyl-2-thiazolyl) bromide-2, 5-diphenyl-2H-tetrazolium)
Fig. 2. (a) UV–visible spectra of Cr(III) complexes 1–8 (1.0 Â 10À4 M) through Fenton-like reaction in PBS (pH 7.4, 10 mM) (Inset: absorbance at 532 nm of samples). (b)
Absorbance at 532 nm of CrCl3, complexes 1 and 8 through Fenton-like reaction in PBS (red), RPMI medium 1640 (green) and HSA (blue). B: Blank group, C: Control group.
PBS/HAS/RPMI medium 1640 act as blank group, Fe(EDTA) act as control group.

156
J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
Fig. 3. (a) The UV–visible spectra of complexes 1–8 (1.0 Â 10À4 M) oxidized by H2O2 (1.0 Â 10À4 M, PBS buffer) in PBS/HSA/RPMI medium 1640 with the addition of
diphenylcarbazide (0.01 g/mL) and concentrated sulfuric acid. (b) Absorbance at 548 nm of UV–visible spectra in Fig. 3a (red: PBS, green: HSA, blue: RPMI medium 1640). (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
assay against MCF-7 cells, and this method was based on the the-
ory that MTT can be reduced by mitochondrial dehydrogenases of
living cells to purple formazan. The formazan formed by living cell
is soluble in DMSO with maximum absorbance at 490 and 570 nm.
The result was given in Fig. 4, and the corresponding ligands were
also employed as a control experiment (Fig. S27). Considering the
impact of DMSO, cell viability of all samples attained to 90%
although cell viability of 1–6 is slightly lower than that of 7 and
8. That’s to say, all complexes behave hypotoxic to cells lines
histology sections of C57 rats treated with Cr(III) compounds
showed normal structures. No adverse effects were found on the
histoarchitecture of hepatocytes after administered acutely. All of
them revealed that high-dose treatment with of Cr(III) compounds
exhibit low toxicity or non-toxic to mice in this period, and this is
accordant with the literature reports [4].
It has been reported by in extension reviews that the toxic
effects of Cr(pic)3 arises only in situations, for example, when the
special contact molecule is present in cell culture, injection into
mammals and animals with simple digestive systems such as fruit
flies); in mammals, the compound degrades in the stomach and is
non-toxic [34].
Å
within 24 h, and the result is well consistent with OH-generation
and Cr(VI)-generation reaction in 1640 or HSA medium. Therefore,
Cr(III) complexes is nearly non-toxic to cell lines and the worry
about Cr(VI)-generation in cells should be ignored [17].
3.6. Anti-hyperglycemic activity study
3.5. Acute toxicity study
To assess the therapeutic effects of synthesized complexes on
diabetes, the hypoglycemic activity of complex 8 on STZ-mice
was examined for 2 months due to it minimum potential hazards
To verify the acute toxicity of these Cr(III) complexes, 1 and 8
was selected as experimental group and CrCl3 was used as a con-
trol group. All mice survived to the end and no significant effects
of the complexes on growth or final mean body mass of mice were
observed (Table S2, Fig. S28) after administration of Cr(III) com-
plexes (10.0 g/kg BW) and DMSO (1%) by gavage, and the
microanatomical features of representational tissue slices were
given as Fig. 5. There was no distinct disparity noticed in body
mass between the experimental group and control group
(Fig. S28 and Table S2), and there seems to be no damage in kidney,
liver and pancreas tissue from the outer appearance (Fig. 5). Kidney
Å
during OH and Cr(VI) generation experiments. The FBG, FINS, TC,
TG, HDL and LDL levels of the lean control mice and STZ-mice were
monitored after oral gavage of CrCl3Á6H2O, 1 and 8 at 0, 4th and 8th
week (Fig. 6). No significant change in physical activities occurred
on the administration of both complex, and all the mice survived to
the end.
FBG (6.07 0.92 mmol/L), FINS (12.36 1.50 lmol/L), TC
(1.92 0.15 mmol/L) and TG (0.55 0.05 mmol/L) levels of mice
in control group are in the normal range, while the mice in diabetic
control group are significantly (p < 0.05) higher than those of the
control group during the experimental process (Table 3). FBG levels
for the diabetic control group is more than 11.1 mmol/L represent-
ing the mice are hyperglycemic. Meanwhile, HFD significantly
affects other parameters for the diabetic control group.
In comparison with diabetic control group, the FBG levels
decreased significantly after administration with CrCl3Á6H2O and
complex
1 after 4 weeks (p < 0.05). FBG levels of the mice
decreased 27.42% and 26.23% for CrCl3 and complex 1 groups after
8 weeks, while complex 8 only led to a slightly decrease (10.26%) of
FBG levels after 8 weeks. Even so, the FBG levels continued to be
much higher than normal mice after treatment with complexes 1
and 8 for 8 weeks (5.60 vs. ꢀ9.42 mmol/L, Fig. 6a).
FINS levels for CrCl3Á6H2O group decreased 12.98% after
4 weeks and the decrease was abolished for 8 weeks (Fig. 6b).
While FINS levels of 1 and 8 treated mice remained no obvious
decrease and they were much higher than that of normal group.
TC levels decreased 18.22% and 1.80% for CrCl3Á6H2O and 1 group
after 8 weeks treatment, respectively. No significant changes were
observed for TG levels in these three Cr(III) complexes group
(Fig. 6c and d). The decrease of FBG, FINS and TC for CrCl3Á6H2O
Fig. 4. Effect of the complexes 1–8 (1.0 Â 10À4 M) on viability of MCF-7 cells (4500),
B: blank, D: DMSO.

J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
157
Fig. 5. Effects of CrCl3, complexes 1 and 8 on kidney, liver and pancreas tissue slices of the mice in acute toxicity studies. Control group: DMSO (1%). Experimental group:
CrCl3, 1 and 8 (10.0 g/kg BW).
Fig. 6. Effect of Cr supplementation on fasting blood glucose (FBG, mmol/L) (a), fasting serum insulin (FINS, mmol/L) (b), Total cholesterol (TC, mmol/L) (c), Triglyceride (TG,
mmol/L) (d), high density lipoprotein (HDL, mmol/L) and low density lipoprotein (LDL, mmol/L) of mice. Rats were supplemented with control buffer, CrCl3, 1 or 8 (1000 mg
Cr/kg body mass) by gavage daily for 8 weeks. #: diabetic group versus normal control (p < 0.05). *: other group versus diabetic group (p < 0.05).

158
J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
Table 3
Alter of body mass, FBG, FINS, TC, TG HDL and LDL levels of control group and the mice followed administration with CrCl3, 1 and 8 carrier (control) by oral gavage during anti-
hyperglycemic activity study. #: diabetic group versus normal control (p < 0.05). *: other group versus diabetic group (p < 0.05).
Week
Normal rats
Diabetic rats
CrCl3 treated rats
1-treated rats
8-treated rats
BM/g
0
4
8
0
4
8
0
4
8
0
4
8
0
4
8
0
4
8
0
4
8
19.24 1.16
22.45 1.55
29.35 1.48
6.07 0.92*
5.54 0.23*
5.60 0.20*
12.36 1.50*
11.45 0.52*
11.25 0.34*
1.92 0.15*
1.74 0.12*
1.92 0.093*
0.55 0.050
0.50 0.043*
0.52 0.026*
1.31 0.11*
1.24 0.092*
1.19 0.086*
0.75 0.064
0.68 0.027*
0.70 0.024*
23.62 2.05
30.5 2.40
24.14 2.14
29.42 1.62
37.89 3.01
12.18 1.61#
9.02 0.28#*
8.84 0.30#*
19.80 0.54#
17.23 0.91#*
17.22 0.86#*
4.28 0.32#
3.77 0.19#
3.50 0.24#*
2.02 0.19
23.39 1.93
29.36 1.96
39.75 1.72
12.77 1.57#
9.91 0.52#
9.42 0.38#
18.78 1.60#
19.64 0.58#*
18.40 0.58#*
4.41 0.74#
3.90 0.26#
4.06 0.23
23.32 1.96
29.51 0.82
39.91 2.23
11.47 1.62#
10.03 0.40#
10.26 0.31#
20.37 1.41#
20.22 0.85#
19.89 1.11#
3.97 0.54#
4.34 0.50#
4.18 0.19#
1.79 0.42
40.82 2.18
11.59 0.39#
10.50 0.52#
10.37 0.35#
19.13 1.21#
21.71 0.52#
21.11 0.66#
4.66 0.64#
4.17 0.32#
4.26 0.23
FBG/mmolÁLÀ1
FINS/l
molÁLÀ1
TC/mmolÁLÀ1
TG/mmolÁLÀ1
HDL/mmolÁLÀ1
LDL/mmolÁLÀ1
1.53 0.22
1.85 0.22
1.92 0.21#
2.04 0.21#
0.90 0.099#
0.80 0.064#
0.75 0.040#
0.87 0.050
0.99 0.064#
1.09 0.062#
1.67 0.20*
1.63 0.14*
0.87 0.062#
0.98 0.059#
0.86 0.049#
0.85 0.095
0.76 0.052*
0.81 0.066*
1.85 0.22#
1.78 0.11#
0.78 0.040#
0.92 0.061#
0.84 0.070#
0.74 0.10
1.81 0.18#
2.02 0.16
0.95 0.16#
0.81 0.053#
0.79 0.069#
0.92 0.19
0.91 0.052#
0.93 0.052#
0.86 0.050#
0.89 0.10
group is bigger than that of 1 and 8 groups. It may be due to the
inert activity of complexes 1 and 8 making Cr(III) difficult to be
decomposed and absorbed.
Moreover, the HDL, LDL levels of mice was also examined in
STZ-mice for 56 days. As seen in Fig. 6e-f, HDL and LDL levels of
all the Cr(III) complexes group remained changeless throughout
the administration. Besides, all drugs have no obvious decrease
on the body mass of the mice (Fig. S29), and this is consistent with
the literature reported previously [8].
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2017.05.041.
References
[1] K. Schwarz, W. Mertz, Arch. Biochem. Biophys. 85 (1959) 292–295.
[2] J.B. Vincent, Biol. Trace Elem. Res. 166 (2015) 7–12.
[3] H. Wang, A. Kruszewski, D.L. Brautigan, Biochemistry 44 (2005) 8167–8175.
[4] A. Levina, T.H. Pham, P.A. Lay, Angew. Chem. Int. Ed. 55 (2016) 8104–8107.
[5] J.B. Vincent, Polyhedron 20 (2001) 1–26.
[6] J.K. Speetjens, R.A. Collins, J.B. Vincent, S.A. Woski, Chem. Res. Toxicol. 12
(1999) 483–487.
4. Conclusion
[7] J.B. Vincent, Met. Ions Life Sci. 13 (2013) 171–198.
[8] M.D. Stout, A. Nyska, B.J. Collins, K.L. Witt, G.E. Kissling, D.E. Malarkey, M.J.
Hooth, Food Chem. Toxicol. 47 (2009) 729–733.
[9] Z. Liu, A. Habtemariam, A.M. Pizarro, S.A. Fletcher, A. Kisova, O. Vrana, L.
Salassa, P.C. Bruijnincx, G.J. Clarkson, V. Brabec, P.J. Sadler, J. Med. Chem. 54
(2011) 3011–3026.
[10] C. Wang, J. Liu, Z. Tian, M. Tian, L. Tian, W. Zhao, Z. Liu, Dalton Trans. (2017).
[11] M.R. Kandadi, M.K. Unnikrishnan, A.K. Warrier, M. Du, J. Ren, N. Sreejayan, J.
Inorg. Biochem. 105 (2011) 58–62.
[12] H.G. Preuss, B. Echard, D. Clouatre, D. Bagchi, N.V. Perricone, J. Inorg. Biochem.
105 (2011) 1344–1349.
[13] S.K. Jain, G. Kahlon, L. Morehead, R. Dhawan, B. Lieblong, T. Stapleton, G.
Caldito, R. Hoeldtke, S.N. Levine, P.F. Bass 3rd, Mol. Nutr. Food Res. 56 (2012)
1333–1341.
[14] S.K. Jain, J.L. Croad, T. Velusamy, J.L. Rains, R. Bull, Mol. Nutr. Food Res. 54
(2010) 1371–1380.
[15] J.B. Vincent, J. Am. Coll. Nutr. 18 (1999) 6–12.
[16] M.M. Abdel-Monem, M.D. Anderson, in: U.S. Patent (Ed.), Zinpro Corporation,
United States, 2007, pp. 1–7.
[17] H.Y. Tang, Q.G. Xiao, H.B. Xu, Y. Zhang, J. Trace Elem. Med. Biol. 29 (2015) 136–
144.
[18] B. Liu, J. Chai, X.Q. Hu, Y.J. Zhang, J.X. Nan, B.S. Yang, Inorg. Chem. Commun. 52
(2015) 27–30.
[19] G.H. Xu, J.F. Ma, Y.Y. Liu, S.L. Li, Acta Crystallogr., Sect. C 62 (2006) m581–
m583.
[20] V. Nair, G.A. Turner, Lipids 19 (1984) 804–805.
[21] I. Mulyani, A. Levina, P.A. Lay, Angew. Chem. Int. Ed. 43 (2004) 4504–4507.
[22] D.M. Steams, W.H. Armsbong, Inorg. Chem. 31 (1992) 5178–5184.
[23] G.M. Sheldrick, Acta Crystallogr. A 64 (2008) 112–122.
[24] J.E.T. Andersen, Anal. Chim. Acta 361 (1998) 125–131.
[25] M. Pettine, T. La noce, A. Liberatori, L. Loreti, Anal. Chim. Acta 209 (1988) 315–
319.
In an attempt to improve the safety and biological activity of Cr
(pic)3, seven new analogues were obtained through pic ligands
with different substituent groups. The structure, Fenton-like reac-
tion, oxidation reaction, cytotoxicity, acute toxicity and hypo-
glycemic activity of all the complexes are evaluated. It was found
that the substituent groups affected the physicochemical activities
Å
of the complexes such as OH- and Cr(VI)-generation in tube. For
Å
example, halogen group complexes generate more OH and Cr(VI)
than complexes with –OH group, however, these differences can
be neglected in cell and animal experiments. This is attributed to
the lower concentration of oxidant but high concentration of bio-
reductants in vivo. This concept was verified by cell MTT assay
and acute toxicity study: no obvious cellular damage and tissue
injury were observed for Cr(pic)3 and its derivatives. On the other
hand, hypoglycemic activity study indicated that Cr(R-pic)3 com-
plexes including Cr(pic)3 have not significant influence than CrCl3-
Á6H2O salt on the blood glucose, serum insulin, total cholesterol,
triglyceride, high density lipoprotein and low density lipoprotein
of diabetic mice. Once again this result questioned the widely use
of Cr(pic)3 as a validity nutrition additives.
Acknowledgements
[26] F. Li, X. Wu, Y. Zou, T. Zhao, M. Zhang, W. Feng, L. Yang, Food Chem. Toxicol. 50
(2012) 1623–1631.
[27] N. Clement, C. Toussaint, G. Rogez, C. Loose, J. Kortus, L. Brelot, S. Choua, S.
Dagorne, P. Turek, R. Welter, Dalton Trans. 39 (2010) 4579–4585.
[28] V.C. Gibson, C. Newton, C. Redshaw, G.A. Solan, A.J.P. White, D.J. Williams,
Dalton Trans. (2003) 4612–4617.
We thank the National Natural Science Foundation of PR China
(No. 21271122, 21571117), Research Project Supported by Shanxi
Scholarship Council of China (2013-018). We thank Scientific
Instrument Center of Shanxi University for the technical assistance.

J. Chai et al. / Inorganica Chimica Acta 466 (2017) 151–159
159
[29] H.G. Raubenheimer, A. du Toit, M. du Toit, J. An, L. van Niekerk, S. Cronje, C.
[32] M. Yamaguchi, J. Phys. Chem. A 115 (2011) 14620–14628.
Esterhuysen, A.M. Crouch, Dalton Trans. (2004) 1173–1180.
[30] Y.Y. Zhang, Q.X. Zhou, Y. Zheng, K. Li, G.Y. Jiang, Y.J. Hou, B.W. Zhang, X.S.
Wang, Inorg. Chem. 55 (2016) 4296–4300.
[33] J.M. Chen, O.J. Hao, Crit. Rev. Environ. Sci. Tec. 28 (2010) 219–251.
[34] J.B. Vincent, Toxicology of chromium(III), in: The Bioinorganic Chemistry of
Chromium, John Wiley & Sons, Ltd, 2012, pp. 195–213.
[31] N. Selvarajan, N.V. Raghavan, J. Phys. Chem. 84 (1980) 2548–2551.