University of Southern California Dissertations and Theses

Reconstruction of a proinsulin-transferrin fusion protein

(USC Thesis Other)

Reconstruction of a proinsulin-transferrin fusion protein

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Reconstruction of A Proinsulin-Transferrin Fusion Protein

by

Yijing Huang

A Thesis Presented to the
FACULTY OF THE USC SCHOOL OF PHARMACY
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(PHARMACEUTICAL SCIENCES)

August 2020

Copyright 2020 Yijing Huang
ii
Acknowledgements
Foremost, I would like to give my greatest gratitude to my advisor, Dr. Bangyan L. Stiles. She has
taught me a lot on scientific research and influenced me much for her rigorousness and hard-
working attitude. I would not have completed my project and this thesis without her selflessly
sharing knowledge and experience. In addition, I would like to thank Dr. Wei-Chiang Shen and
Dr. Lina He for their support and insightful instruction on this project. I would also like to thank
my committee members, Dr. Curtis Okamoto and Dr. Jennica Zaro, for their time and suggestions
on this thesis.

In the past two years of my study at USC, I have luckily met with many senior students who have
given me invaluable kindness and help. They are Dr. Sean Wang, Dr. Joshua Chen, Dr. Jingyu
“Sophia” Chen, Taojian “Michael” Tu, and Lulu Chen. Furthermore, I would like to thank many
classmates and people at SOP for spending time with me and bringing me happiness. I would not
have enjoyed the life at USC without them.

Last but not least, I would like to express my deep gratitude to my dearest parents for their love,
understandings and huge priceless support on my study in the US. They have given everything to
me. In addition, my special thanks go to my other family members and many friends in China for
their encouragements and emotional support.
iii
Table of Contents
Acknowledgements ......................................................................................................................... ii
List of Tables .................................................................................................................................. v
List of Figures ................................................................................................................................ vi
Abbreviations ................................................................................................................................ vii
Abstract .......................................................................................................................................... ix
CHAPTER I. Introduction .............................................................................................................. 1
I. Diabetes and Insulin Discovery ............................................................................................... 1
II. Biosynthesis of Insulin ........................................................................................................... 2
III. History of Therapy Using Insulin Mimetics ......................................................................... 2
III.1 Recombinant Human Insulin .......................................................................................... 5
III.2 Insulin Mimetics ............................................................................................................. 7
III.2.1 Mealtime Insulin Mimetics ...................................................................................... 7
III.2.1.1 Background ....................................................................................................... 7
III.2.1.2 Short-Acting Insulin Analogues ....................................................................... 9
III.2.2 Basal Insulin Mimetics .......................................................................................... 14
III.2.2.1 Background ..................................................................................................... 14
III2.2.2 Long-Acting Insulin Analogues ....................................................................... 15
III.2.3 Production of Insulin Analogues ........................................................................... 19
III.2.4 Future Development of Insulin Mimetics .............................................................. 19
IV. Proinsulin as A Hepatic Targeting Insulin Molecule ......................................................... 20
V. Transferrin and Its Receptor-Mediated Endocytosis and Recycling Pathway ..................... 22
CHAPTER II ．Plasmid Design and Cloning ............................................................................... 25
I. ProINS-hTf Fusion Protein .................................................................................................... 25
II. ProINS-mTf Plasmid Design ............................................................................................... 26
III. Results ................................................................................................................................. 29
III.1 PCR to generate the insert............................................................................................. 29
III.2 Generating ProINS as receiving plasmid vector ........................................................... 30
III.3 Ligation ......................................................................................................................... 30
CHAPTER III. Expression and Characterization of ProINS-mTf Recombinant Fusion Protein . 36
iv
CHAPTER IV. Discussion and Conclusion.................................................................................. 39
I. Discussion .............................................................................................................................. 39
II. Conclusion ............................................................................................................................ 42
Materials and Methods .................................................................................................................. 43
References ..................................................................................................................................... 47

v
List of Tables
Table 1. Application of Transferrin in Fusion Proteins ................................................................ 24
Table 2. Sequence comparison and following translated amino acids among the record from
website, PCR templates and plasmid No.7 ................................................................................... 35

vi
List of Figures
Figure 1. Biosynthesis of insulin .................................................................................................... 3
Figure 2. Development of insulin therapy and structures of insulin from different species ........... 4
Figure 3. Comparison of amino acid sequence among human insulin, human IGF-1 and ILisp . 11
Figure 4. Structure of insulin analogues compared with hINS. .................................................... 18
Figure 5. Transferrin receptor-mediated endocytosis and recycling pathway .............................. 23
Figure 6. Plasmid map as indicated .............................................................................................. 25
Figure 7. Pairwise amino acid sequence alignment between Tf from human and mouse (from
EMBOSS) ..................................................................................................................................... 27
Figure 8. Study design .................................................................................................................. 28
Figure 9. Results of DNA agarose gel .......................................................................................... 31
Figure 10. Plasmids extracted from colonies ................................................................................ 34
Figure 11. Western blot result of concentrated cell media after transfection ............................... 37

vii
Abbreviations
CSII: Continuous Subcutaneous Insulin Infusion
E. coli: Escherichia coli
ER: Endoplasmic Reticulum
Fig.: Figure
IAsp: Insulin Aspart
IDeg: Insulin Degludec
IDet: Insulin Detemir
IGF1-R: Insulin Growth Factor 1 Receptor
IGlar: Insulin Glargine
IGlu: Insulin Glulisine
ILisp: Insulin Lispro
INS/ProINS/rhINS: Insulin/Proinsulin/ Recombinant Human Insulin
IR: Insulin Receptor
Lipo3000/2000: Lipofectamine 3000/2000
NPH: Neutral Protamine Hagedorn
PCR: Polymerase Chain Reaction
PEI: Polyethylenimine
S. cerevisiae: Saccharomyces cerevisiae
viii
Tf/hTf/mTf: Transferrin/Human Transferrin/Mouse Transferrin
-gal: -Galactosidase

ix
Abstract
Insulin mimetics has been developed since the discovery of insulin in 1921. The ultimate goal is
to normalize blood glucose level in patients with diabetes. Previous study has designed and
characterized a human proinsulin-transferrin (ProINS-Tf) recombinant fusion protein, which is a
long-term insulin prodrug with liver selectivity. However, it showed immune reaction
counteracting the effect of the human fusion protein in mouse models, prohibiting further in vivo
studies. To overcome this immunogenicity, the present study is designed to replace human
transferrin with mouse transferrin through subcloning. The resultant plasmid was successfully
reconstructed and the protein produced in HEK293 cells was identified by immunoblotting. This
thesis work will allow further studies to validate ProINS-Tf as a potential therapy for treating
diabetes in patients.

1
CHAPTER I. Introduction
I. Diabetes and Insulin Discovery
Diabetes is a metabolic disease with a high level of circulating glucose. According to the World
Health Organization, ~1.6 million patients deceased due to diabetes directly in 2016. In 2019, 10.9%
of adults in the United States have had diabetes. Patients with diabetes can develop various
complications, such as blindness, kidney failure, and heart diseases, which affect life quality of
millions of people. Diabetes is usually caused by the failure to produce insulin or to respond to
insulin, which are called type 1 and type 2 diabetes, respectively.

Insulin produced by  cells in the pancreas is a well-studied polypeptide hormone composed of
two chains to regulate glycemic level in the body. At the beginning of the 20th century before
insulin was first isolated, extracts of animal pancreas were found able to lower blood glucose level
in vivo (Rosenfeld, 2002). In 1921, Frederick G. Banting, Charles Best and J.J.R. Macleod found
a way to remove the digestive enzymes during the extraction to successfully keep the islet cells
alive in the pancreas. J. B. Collip further improved the extraction procedure using acid alcohol
(Bliss, 1993). Their works led to the successful discovery of the pancreatic hormone insulin. Since
then, insulin has been investigated as treatment for diabetes.
2
II. Biosynthesis of Insulin
In Figure 1 (Fig.1), after transcription of INS gene, preproinsulin, a single-chain precursor to
insulin, is synthesized by pancreatic β-cells. Upon translocation to the ER membrane,
preproinsulin is, thereafter, cleaved into proinsulin and a signal peptide (Tsuchiya et al., 2018). In
the ER, the reduced and unfolded proinsulin is typically oxidatively folded via three disulfide
bonds, which is catalyzed by protein disulfide isomerase, for stability and bioactivity (Weiss,
2009). The folded soluble proinsulin transits through the Golgi apparatus and is sorted into
immature secretory granules (Weiss, 2009). The proinsulin is then further converted into insulin
through cleavage of C-peptide, a connecting polypeptide between A- and B-domains (Huang and
Arvan, 1994). Insulin is stored as hexamers stabilized by Zn2+ in mature secretory granules, in
which C-peptide is stored together in equimolar amounts (Yosten et al., 2014). In response to high
glucose conditions, insulin and C-peptide in granules are together secreted into the portal vein to
enter the liver, where insulin hexamers dissociate into monomers (Weiss, 2009).

III. History of Therapy Using Insulin Mimetics
Insulin extracted and produced from animals was first used to treat diabetic patients (Fig.2). To
reduce the number of injections per day, protamine and zinc were added to reduce the solubility
of insulin products to prolong the action, leading to the development of long-acting insulin
products (Bliss, 1993). Physiologically, insulin can be stored in mature secretory granules in the
form of a zinc-containing hexamer insulin before secretion (Emdin et al., 1980). Native insulin
3
molecules tend to form dimers. With the help of zinc ions, three dimers form one hexamer. The
aggregation reduces the solubility of insulin and prevents the hexameric insulin from contacting
cell membrane where enzymes degrading insulin are located (Emdin et al., 1980). Therefore, zinc
increases stability of insulin formulation.

However, long-term exposure to animal-sourced insulin, an exogenous polypeptide, resulted in
immune reactions (Kahn and Rosenthal, 1979). Levels of antibodies to insulin in the blood could
reach the threshold of detection in almost all diabetic patients when treatment of insulin were
sustained for two months (Kahn and Rosenthal, 1979). These antibodies can interact with the
exogenous insulin, rendering a delay or a loss of bioactivity of these therapies. When exposed to
animal insulin chronically, some diabetic patients develop patterns of host immune response,
Figure 1. Biosynthesis of insulin
“Pre” in preproinsulin represents a signal peptide that allows the molecule to enter the
secretory pathway. “B”, “C”, and “A” refer to the indicated domain. Mature insulin
consists of two chain, A-chain and B-chain, which are linked through three disulfide
bonds (the dark line in the insulin structure). The “C” domain is cleaved and released
during post-translational processing.
4

Figure 2. Development of insulin therapy and structures of insulin from different species
(A) Timeline of insulin therapy development. (B-D) Comparison of human, bovine and
porcine insulin. Porcine insulin has a different amino acid than human insulin while bovine
insulin has three.
5
which is referred to as “insulin allergy” (Kahn and Rosenthal, 1979). Insulin allergy is the most
common consequence in patients treated with insulin. In addition, “immunological insulin
resistance” is also observed in some patients. Animal insulin induce the immune system in type 1
patients and elevated level of insulin IgG. Occasionally, this IgG sequesters insulin given in
therapy, preventing it from interacting with the receptor. As a result, these patients would require
more insulin than normal, which is referred to as immunological resistance to insulin.

To overcome the immunogenicity problems associated with foreign antigen, a semi-synthesis
method was developed by modifying porcine insulin as porcine insulin differs from human insulin
with only one amino acid, whereas three amino acid difference is observed for bovine insulin
(Fig.2C). Novo Nordisk A/S established a procedure to produce this semisynthetic humanized
porcine insulin in the 1970s (Ladisch and Kohlmann, 1992).

III.1 Recombinant Human Insulin
In the late 1970s, genetic engineering methods were developed to produce recombinant human
insulin (rhINS) in a large quantity. In 1978, David Goeddel produced the first recombinant DNA
human insulin, which was then commercially available in 1982 (Quianzon and Cheikh, 2012). In
this approach, the human insulin A- and B-chains were separately cloned into pBR322 containing
-galactosidase (-gal) (Ladisch and Kohlmann, 1992). The -gal chimeric polypeptides
containing A- or B- chains were expressed, amplified and obtained in Escherichia coli (E. coli).
6
The two chains were then released after a chemical cleavage of -gal. The rhINS was produced
through chemical linkage of the two chains and then purified.

Compared to the two-chain method, Kroeff established a less complex way of rhINS production
where proinsulin, a one-chain-precursor to insulin, was produced first in E. coli, followed a
conversion to rhINS (Kroeff et al., 1989; Quianzon and Cheikh, 2012). In this process, A- and B-
chains are linked through a connecting peptide (C-peptide) in proinsulin, as is the case with
endogenous insulin synthesis. The proinsulin was obtained via reverse transcription of proinsulin
mRNA, cloning the cDNA into a -gal-containing plasmid and expression in E. coli. The resulting
proinsulin was then converted to rhINS through oxidative folding with formation of disulfide
bonds, and cleavage of C-peptide. Besides bacteria (E. coli), yeast (like Saccharomyces cerevisiae
or S. cerevisiae) can also be used to produce proinsulin (Tottrup and Carlsen, 1990). In addition,
the cDNA sequences of insulin A- and B- chains can also be fused directly or via a polypeptide
linker in the plasmid construct (Baeshen et al., 2014). Proinsulin can be produced either as a
secreted or intracellular protein. The secreted proinsulin is more readily purified than the
intracellular one as there are less proteins in the cell media comparatively (Ladisch and Kohlmann,
1992). Compared to animal insulin, rhINS are less immunogenic, leading to rare occurrence of
lipoatrophy and insulin allergy in patients (Fineberg et al., 1983; Schernthaner, 1993).

7
III.2 Insulin Mimetics
While rhINS solves the immunogenicity problem of animal-sourced insulin, it fails to consistently
normalize blood glucose level due to the physiological differences between rhINS and endogenous
insulin. During fed state, insulin is secreted from the pancreatic  cells at a relatively large quantity
within a short time upon food absorption into the circulation. Mealtime insulin acts on the muscle
and adipose tissue as well as the liver, to simulate glucose uptake from the circulation and suppress
glucose output to the blood, leading to a decrease in blood glucose level. During fasting state,
insulin secretion is lower than mealtime and thus inhibition of hepatic glucose release by insulin
is diminished. Together with glucagon, a hormone that increases the blood glucose during fasting,
insulin maintains blood glucose level within a narrow range, 70-100 mg/dL. To achieve this
narrow range, two types of insulin products mimicking mealtime and non-mealtime secretion of
insulin have been developed to meet different physiological conditions.

III.2.1 Mealtime Insulin Mimetics
III.2.1.1 Background
Limitation of rhINS. For insulin-dependent diabetes, pre-prandial insulin products with fast onset
of action are necessary to supply mealtime insulin. The rhINS, however, shows a lag-phase
following injection in achieving its glucose lowering effect and fails to normalize the glycemic
level in a timely manner. Insulin self-association is a key limiting factor that prevents the rhINS
from achieving this fast glucose-lowering effects. In the formulation of rhINS (as well as porcine
8
insulin), zinc and phenolic preservatives are used to enhance stability in a typical insulin
formulation (Derewenda et al., 1989; Richards et al., 1998). While these additions increase the
stability of rhINS, they also lead to reduced bioavailability before injection (Bolli et al., 1999).
The insulin hexamer, the major form of insulin in the zinc and phenolic formulated insulin therapy,
needs to dissociate into dimeric or monomeric insulin before absorption to the circulatory system
following subcutaneous injection. Limited by the rate of dissociation, absorption of insulin from
the injection site is often too slow to provide a sufficiently high concentration of insulin in the
blood during a meal, preventing it from lowering the high blood glucose level 1-2 hours after a
meal (Dimitriadis and Gerich, 1983). Moreover, as continued absorption from the injection site
can last for a few hours, an unusual hyperinsulinemia in the blood can occur in patients 4-6 hours
after a meal, resulting in unwanted hypoglycemia (Dimitriadis and Gerich, 1983). Hypoglycemia
can be accompanied by complications, including myocardial infarction, retinopathy, brain diseases,
neurocognitive dysfunction as well as unpleasant quality of daily life in diabetic patients (Kalra et
al., 2013). Therefore, considerable interest was focused on developing a stable insulin product that
can mimic the mealtime endogenous insulin to overcome the shortcomings of rhINS.

Short-acting Monomeric Insulin as Solution. Unlike hexameric insulin, monomeric insulin can be
quickly absorbed to the blood and is also the active form of insulin. Interference of insulin self-
association, therefore, is the key in the development of short-acting insulin products. The effort
includes both formulation and structure alterations of rhINS. Zinc-free rhINS formulation, for
9
example, can significantly reduce insulin hexamerization, and thereby reduce the absorption time.
However, zinc-free formulation has some limitations as Zn2+ is useful to stabilize pharmaceutical
insulin. Other than zinc, the phenolic preservatives can also bind to insulin and further promotes
insulin self-association (Derewenda et al., 1989). In the absence of zinc, self-association and
aggregation are also natural characteristics of insulin. These processes are dependent on the pH
and concentration of insulin (Attri et al., 2010). At a millimolar concentration, which is
significantly higher than the physiological insulin concentration of normally less than 1 nM, zinc
is not necessary for insulin hexamerization (Attri et al., 2010; Hansen, 1991). Therefore, zinc-free
insulin preparation is not an ideal solution and structure modifications via recombinant DNA
technology, on the other hand, are developed to produce insulins that are stable as zinc-containing
hexamers in the formulation and dissociate into monomers upon injection at a much faster rate
than rhINS.

III.2.1.2 Short-Acting Insulin Analogues
Insulin dimerization is the basis for hexamerization. The amino acid substitution for human insulin
is, thereafter, developed based on the understandings for how molecular interactions occur
between monomers and between dimers. These insulin products are collectively termed insulin
analogues.

10
Most of the B-chain residues of insulin are involved in dimerization and hexamerization (Brange
et al., 1990). The N-terminus of the B-chain is believed to contribute to metal-binding and the C-
terminus is essential in self-association and receptor-binding (Ciszak et al., 1995). J. Brange
constructed several monomeric insulin analogues by single amino-acid substitutions that did not
self-associate without impairing bioactivity. These analogues showed a significantly higher rate
(2-3 times) of absorption following subcutaneous injection than rapid-acting rhINS at that time
(Brange et al., 1988). For example, as B9 (Ser) and B13 (Glu) residues are close to each other in
dimers and hexamers, replacement of uncharged Ser at B9 with negatively charged Glu introduced
charge repulsion to B13 and destabilized the association (Brange et al., 1988). In addition to charge
repulsion, increased steric hindrance between subunits through the replacement of Val with Ile at
B12 can also interfere with the association (Brange et al., 1990). Third, when hydrophobic residues
involved in the association (Tyr at B16, Leu at B17, and Tyr at B26) are changed to a hydrophilic
residue (Glu), hydrophobic interfaces in dimers and hexamers would be unstable and the
association can be counteracted (Brange et al., 1990). Moreover, since His at B10 participates in
binding with zinc ion, mutation at B10 can weaken metal-binding and hexamerization (Brange et
al., 1990).

The first commercial short-acting insulin analogue is insulin lispro (ILisp), available in the United
States in 1996. Pro at B28 and Lys at B29 in human insulin are inverted in ILisp (Fig.4). The

11
B-domain/chain
IGF-1
(1-29)
G P E T L C G A E L V D A L Q F V C G D R G F Y F N K P T
hINS F V N Q H L C G S H L V E A L Y L V C G E R G F F Y T P K T
ILisp
(1-30)
F V N Q H L C G S H L V E A L Y L V C G E R G F F Y T K P T
C-domain
IGF-1
(30-41)
G Y G S S S R R A P Q T
A-domain/chain
IGF-1
(42-62)
G I V D E C C F R S C D L R R L E M Y C A

hINS G I V E Q C C T S I C S L Y Q L E N Y C N
ILisp
(1-20)
G I V E Q C C T S I C S L Y Q L E N Y C N
D-domain
IGF-1
(63-70)
P L K P A K S A
Figure 3. Comparison of amino acid sequence among human insulin, human IGF-1 and ILisp
Amino acids shared by human insulin, IGF-1 and ILisp are indicated by green shading. Lys-Pro
vs. Pro-Lys sequence differences are shaded by blue color.

design of ILisp is inspired by the structure insulin-like growth factor-I (IGF-1, Fig.3), which is
highly homologous with insulin but does not self-associate (Bolli et al., 1999). Compared with
human insulin, the C terminal of IGF-1 B-domain has Lys-Pro sequence at position 27 and 28,
which is reversed in insulin B-chain (Fig 3). This difference causes the loss of self-association in
IGF-1 (Brems et al., 1992), whereas the C-terminus of insulin B-chain is crucial in insulin
dimerization. Given these findings, ILisp was produced in E. coli through recombinant DNA
technology. Similar to IGF-1, ILisp has Lys-Pro sequence at position 28 and 29 of B-chain rather
than Pro-Lys (Fig 3 and 4). As a result, ILisp displayed reduced dimerization, leading to a similar
12
absorption profile as a monomeric rhINS (Radziuk et al., 1997). The loss of the association is
mainly attributed to the mutation at B28 of insulin. Lys at B29 is involved in dimerization but not
of equal importance as Pro at B28 (Brems et al., 1992). Pro at B28 is found to form intermolecular
non-bonded contacts with B20, B21and B23, which is critical in dimerization (Baker et al., 1988;
Brems et al., 1992). Substitutions of Pro-Lys by Lys-Pro at B28 and B29 also lead to
conformational changes and result in the impairment or loss of these interactions between
monomers.

These alterations drastically weaken interactions of self-association. As a result, the ILisp
hexamerization is only possible with the presence of zinc ions and phenolic molecules in the vial.
Therefore, ILisp in the pharmaceutical formulation can still exist as discrete hexamers to ensure
its stability (Ciszak et al., 1995). When injected, the ILisp hexamers are less stable than the rhINS
hexamers. ILisp can rapidly dissociate into monomeric insulin due to the diffusion of phenolic
molecules and zinc ions to the surrounding tissue (Ciszak et al., 1995). The majority of ILisp in
the subcutaneous tissue can be more readily absorbed than rhINS. Upon absorption into the blood,
ILisp can act as a monomer and exerts a rapid hypoglycemic effect. These properties allow ILisp
to be short acting analogue of rhINS in vivo (Radziuk et al., 1997).

Insulin aspart (IAsp), the second introduced monomeric analogue approved in the US in 2000, is
designed based on the replacement of Pro at B28 with negatively charged Asp (Fig.4). Zinc ions
13
and phenols are also used in the IAsp formulation. Therefore, IAsp can maintain its hexameric
state in the vial and achieve quick dissociation into monomers upon injection. Overall, ILisp and
IAsp are indistinguishable from each other in regard to their absorption behavior, onset time, half-
life and hypoglycemic effect in type 1 diabetic patients (Homko et al., 2003; Plank et al., 2002).

Insulin glulisine (IGlu), approved by FDA in 2004 in the US, is a new rapid-acting insulin analogue
formulated without zinc. Asn at B3 is replaced with Lys, and Lys at B29 is replaced with Glu
(Fig.4). The exchange at B3 does not affect the hexamerization induced by Zn2+, but results in the
loss of the interaction with phenols (Becker et al., 2007; Becker and Frick, 2008). The exchange
at B29 allows more dimerization than that at B28. Furthermore, Glu at B29 can stabilize the
monomer by interaction with Gly at A1 (Becker, 2007). Together, these structure and formulation
changes allow IGlu to dimerize, but reduce its denaturation when dissociated as a monomer
(Becker, 2007). Polysorbate 20 is a surfactant that can prevent denaturation of insulin monomer
and thereby stabilize IGlu (Becker, 2007). The formulation with polysorbate 20 instead of zinc
also improved the stability of IGlu in the vial. Upon injection, IGlu can rapidly dissociate into
monomers. IGlu exerted a faster onset of action than ILisp in non-diabetic subjects, although the
hypoglycemic effect was not different between the two (Heise et al., 2007).

All three rapid-acting insulin analogues, ILisp, IAsp and IGlu exhibit similar pharmacological
effect with limited toxicity (van Bon et al., 2011). In terms of their immunogenicity, none of them
14
are likely to induce immune reactions in patients (Fineberg et al., 1996; Ilag et al., 2016; Zwickl
et al., 1995). Besides metabolic effects, insulin has a relatively low mitogenic potency (Castillo et
al., 2004). A potential mitogenicity is proposed as activation of IGF-1 receptor (IGF1-R) and
prolonged insulin interaction with the insulin receptor (IR) can both induce cell growth effect
(Hansen et al., 1996). Experimentally, ILisp displayed similar or slightly less affinity for the IR,
and moderately more affinity for the IGF1-R when compared with rhINS (Slieker et al., 1997).
However, this slightly increased affinity of ILisp with IGF1-R did not lead to an increase in
stimulating growth of human mammary epithelial cells (HMEC) when compared to rhINS (Slieker
et al., 1997). Comparatively, IAsp binds to both IR and IGF1-R with similar affinity as insulin
(Hansen et al., 1996). The mitogenic effect of IAsp was investigated in human osteosarcoma cells
(Saos/B10), and found to be slightly less potent than rhINS (Kurtzhals et al., 2000). IGlu binds to
IR with a slightly less affinity as rhINS, and to IGF1-R with ~4 times lower. IGlu and rhINS were
not different in stimulating growth of cardiac K6 myoblasts (Stammberger et al., 2006).

III.2.2 Basal Insulin Mimetics
III.2.2.1 Background
In normal physiology, glucose homoeostasis during fasting state and sleep is maintained by a slow
and steady secretion of insulin. Therefore, besides mealtime insulin supply, another goal for insulin
therapy aims to provide diabetic patients with insulin at a low and stable concentration to maintain
a normal glucose level throughout the day. An ideal insulin supply should be a reproducible single
15
dose daily without resulting in hypoglycemia (Barnett, 2003). To achieve this goal, continuous
subcutaneous insulin infusion (CSII), such as an implanted insulin pump has been considered. CSII
resembles the action of physiological insulin, but requires careful monitoring of patients and
professional patient education (American Diabetes Association, 2004). This therapy is currently
largely restricted to type 1 diabetic patients (Pickup and Keen, 2002).

Long-acting insulin analogues based on modification of insulin structure are more practical than
CSII. There are two strategies in the design. One is to prolong the time of absorption from the
injection site. For example, if insulin hexamers are more stable, it can be expected that the
absorption from the injection site would be slowed down. Another one is to find some carrier that
has a long time in the circulation, like serum protein, to extend the action time of insulin products.

III2.2.2 Long-Acting Insulin Analogues
Insulin glargine (IGlar), the first long-acting insulin analogue, was approved in the US in 2001.
Compared to human insulin, Asn is replaced with Gly at A21 in IGlar (Fig.4), and two positively
charged arginine molecules are added to position 30 on insulin B-chain, which changes the
isoelectric point (pI) to 6.7. IGlar (pI = 6.7) is positively charged in an acidic formulation (pH=4),
whereas IGlar is nearly uncharged in a neutral environment. Since molecular solubility is
positively correlated with the extent of charge, IGlar is soluble in the formulation (pH=4) and
becomes insoluble in the subcutaneous tissues (pH = 7) (Bolli et al., 1999). Asn at A21 in human
16
insulin is susceptible to deamidation in the acid solution. The Gly modification in IGlar neutralizes
the molecule and stabilizes it. Due to the decreased solubility, IGlar then can be absorbed slowly
at a relatively constant rate from the subcutaneous tissue. Serving as a depot, IGlar provides a basis
for a long-acting profile that mimics the secretion of basal insulin following injection. The action
profile of IGlar resembles closely that of CSII, where there is a peak that is followed by a declining
action (Bolli and Owens, 2000).

IGlar is developed based on the delayed absorption from the subcutaneous tissue. An alternative
method to achieve a slow absorption is acylation of the molecule, which increases its binding to
albumin (Bolli and Owens, 2000). Albumin is a serum protein made by liver to bind and transport
ions, fatty acids, hormones and some other substances in the bloodstream. Albumin is also known
to bind and transport exogenous substances, like drugs. Molecules bound to albumin exist in
interstitial fluid and plasma cannot take action in the body until they are released. Binding to
albumin also increases the retention time in the bloodstream.

Insulin detemir (IDet), approved by the FDA in 2005 in the US, is a long-acting insulin analogue
with addition of a 14-carbon fatty acyl chain at B29 and omission of threonine at B30 (Fig.4). In
a study of disappearance rate from injection site and plasma profile in vivo, it was found that IDet
prolongs residence time in the injection depot through dihexamerization and albumin binding in
the injection depot, followed by the retention in circulation (Havelund et al., 2004).
17

Besides prolonged duration of action, IDet is found to cause less weight gain, a common side effect
of NPH insulin (Neutral Protamine Hagedorn insulin). This unintended benefit is attributed to the
passive liver-targeting effect of IDet (De Leeuw et al., 2005; Haak et al., 2005; Hordern et al.,
2005). The liver selectivity may be caused by albumin binding. Only IDet free of albumin can
cross the capillary wall and enter peripheral tissues, like muscle and adipose tissues. However,
even bound to albumin, IDet can enter the liver due to the fenestrated endothelial in the sinusoid,
leading to a passive accumulation of these large molecules in the liver. IDet also has an effect on
the central nervous system (CNS) that may be responsible for its anorexigenic response in patients
(Hallschmid et al., 2010). Therefore, albumin binding, preferential activity in brain and liver are
characteristics of IDet that give it the advantage of being the more advanced insulin replacement
therapy (Tibaldi, 2007).

Insulin Degludec (IDeg), approved by the FDA in September 2015, is the newest long acting
insulin analogue (Fig.4). In IDeg, Thr at B30 is removed and a 16-carbon fatty diacid chain is
attached to Lys at B29 via a glutamic acid linker. Formulated with Zn2+ and phenols, IDeg can
only form dihexamers in the vial as the presence of phenolic molecules in the formulation
interferes with the contact of side chains between IDeg monomers (Jonassen et al., 2012). After
subcutaneous injection, as phenolic molecules diffuse away, IDeg can further form soluble
multihexamers (Herring et al., 2014; Meneghini and Miranda-Palma, 2012). As Zn2+ is slowly
18
hINS

ILisp

IAsp

IGlu

IGlar

IDet

IDeg

Figure 4.
Structure of
insulin analogues
compared with
hINS.
Different amino
acids are indicated
by a different font
color (orange vs.
white). The
number indicates
the location of the
chain.

19
diffused away, IDeg monomers are slowly dissociated from the multihexamers and absorbed
(Jonassen et al., 2012). IDeg is an ultra-long-acting insulin analogue with half-life more than 24
hrs (Wang et al., 2012). IDeg can also bind to albumin when it enters the circulation, but it only
contributes minimally to the prolonged action (Kalra, 2013). IDeg has the potential to be
hepatospecific due to the large size and albumin-binding. However, no data demonstrates its
hepatospecificity. In addition, IDeg displays four times lower day-to-day within-subject variability
than IGlar when comparing their glycemic control abilities (Haahr and Heise, 2014).

III.2.3 Production of Insulin Analogues
ILisp, IGlu and IGlar are commercially produced utilizing E. coli. Based on the proinsulin method
to produce rhINS, pro-ILisp, a linear precursor of ILisp, is first expressed in E. coli. After
extraction and enzyme treatment (trypsin and carboxypeptidase), pro-lispro is cleaved into ILisp
(Baeshen et al., 2014). Production of IGlu and IGlar in E. coli is similar as ILisp. IAsp, IDeg and
IDet are produced in S. cerevisiae (yeast) using similar process. For example, pro-IAsp is
expressed and secreted out of the cells. Pro-IAsp is purified from the culture media and excised
into IAsp.

III.2.4 Future Development of Insulin Mimetics
For the rapid-acting analogs, research efforts are focused on developing molecules that are
absorbed into the bloodstream quickly after injection in order to normalize mealtime glycemic
20
level (Tibaldi, 2014). The goal of developing long-acting insulin therapy is to mimic the action of
physiological insulin as closely as possible. Due to the delivery route (subcutaneous injection),
these insulin mimetics enter the circulation system before they reach the liver and peripheral
tissues. However, physiologically, insulin produced by the pancreas is directly secreted into the
portal vein to enter the liver. To reach the concentration of insulin in the liver to take effect after
subcutaneous injection, overexposure of insulin in peripheral tissues cannot be avoided (Tibaldi,
2014), resulting in hypoglycemia and related adverse effects. Although the level of hypoglycemia
caused by exogenous insulin is decreased with long acting insulin, improvement is needed to
achieve a liver targeting insulin therapy.

IV. Proinsulin as A Hepatic Targeting Insulin Molecule
Physiologically, the liver extracts ~60% of total insulin, and hepatocytes are, therefore, exposed
to insulin at much higher concentrations than the peripheral tissues (Hordern et al., 2005). As a
result, it is generally accepted that insulin therapy with hepatoselectivity has potential to better
control glycemia with less variability in patients with diabetes.

Proinsulin, the single-chain precursor of insulin, has been reported to have a preferential effect on
hepatic glucose output than peripheral glucose metabolism, which means proinsulin is
hepatospecific (Galloway et al., 1992; Glauber et al., 1986; Revers et al., 1984). The
hepatospecificity of proinsulin, although not fully understood, is potentially related to its size.
21
Proinsulin is larger than insulin regarding molecular size and thereby diffuses more slowly into
peripheral tissues from the systemic circulation than insulin (Herring et al., 2014). Moreover, in a
study of covalent insulin dimers, it was demonstrated that insulin analogues with increased
molecular size would have less access to peripheral tissues compared to the liver with fenestrated
endothelia, leading to a greater impact on glucose output than glucose disposal (Shojaee-Moradie
et al., 1995). IDet described above also shows hepatic preference. These results suggest that the
larger molecular size may make proinsulin a preferred liver specific insulin therapy. Studies with
proinsulin has demonstrated a relatively longer half-life than insulin. Compared to insulin (t1/2 = 5
min), the disappearance time of endogenous serum proinsulin (t1/2 = 17 min) is found to be longer
in patients following  cell adenoma cut off (Starr and Rubenstein, 1974). In addition, it was
observed that the action of subcutaneous proinsulin has significantly longer acting time than
insulin in vivo (Schatz and Ammermann, 1988).

Proinsulin, however, shows a relatively low bioactivity compared to insulin. Proinsulin is reported
to be ~8% as effective as insulin in stimulating peripheral glucose disposal and ~12% as
biologically active as insulin in suppressing hepatic glucose production (Bergenstal et al., 1984;
Revers et al., 1984). To utilize the larger molecular size and hepatic targeting property of proinsulin,
approaches are needed to improve pharmacological activity and pharmacokinetics properties of
the proinsulin molecule.

22
V. Transferrin and Its Receptor-Mediated Endocytosis and Recycling Pathway
Transferrins are a family of proteins, including transferrin, lactoferrin, and ovotransferrin.
Transferrin (Tf) is a large serum protein responsible for iron transport from the GI tract, the
reticuloendothelial systems (RES) and liver parenchymal cells to proliferative tissues throughout
the body (Kim et al., 2010). The size of Tf (~76 – 81kDa) depends on the extent of glycosylation
(Brandsma et al., 2011). Tf has a long half-life in the circulation in humans, which is generally
~15 days but reduced to ~8 days due to glycosylation (Arndt, 2001; Awai and Brown, 1963; Stibler,
1991).

Tf receptors are ubiquitously distributed in the body. The primary route of cellular iron uptake is
through Tf receptor-mediated endocytosis pathway (Brandsma et al., 2011). Figure 5 shows that
Tf carrying iron can interact with Tf receptor on the cell membrane, leading to the formation of a
complex of Tf and Tf receptor. The complex is endocytosed into the cell as a vesicle, followed by
a decrease in pH of the vesicle to 5.5 or less (Dautry-Varsat et al., 1983). The acidic environment
in the vesicle causes the dissociation of Tf and iron, whereas Tf is still bound to Tf receptor. Apo-
Tf, Tf free of iron, is then recycled back with Tf receptor to the cell surface, upon which the apo-
Tf is dissociated from the complex and released outside the cell. During this process, the change
of pH plays a key role in the association and dissociation of Tf. Apo-Tf can bind to Tf receptor in
an acidic environment with similar affinity as does iron-bearing Tf under pH of ~7.2 (Dautry-
Varsat et al., 1983). Apo-Tf also displays a faster rate of dissociation and association at a pH of
23
~7.2 (Klausner et al., 1983), indicating there is limited competition between iron-bearing Tf and
apo-Tf regarding binding to Tf receptor on the cell surface at a pH ~7.2. Moreover, during the
endocytosis and recycling pathway, some of the endosome containing the complex of Tf and Tf
receptor is generally believed to be connected with secretory protein pathway in the trans-Golgi
network (Stein and Sussman, 1986; Stoorvogel et al., 1988). Therefore, enzymes activating
proteins in the secretory pathway may have access to the complex of Tf and its receptor.
This long plasma half-life of Tf makes it a preferred drug delivery vehicle for pharmaceutical
development. As recombinant DNA technology is developed, Tf has been developed in fusion
Figure 5. Transferrin receptor-mediated endocytosis and recycling pathway
Tf carrying iron can bind to TfR and form a complex to be endocytosed. Unloading iron within
the cell, apo-Tf remains bound to the receptor and recycles back to the membrane, followed by
dissociation from the receptor.

Tf
Iron
Tf receptor
Endosome
cell
membrane
cytoplasm
pH<=5.5
pH~7.2
24
protein engineering (Table 1) as a targeting moiety for protein delivery (Shin et al., 1995), a carrier
moiety for oral protein delivery due to the expression of Tf receptors in the GI tract (Bai et al.,
2005), a radiopharmaceutical for imaging tumor angiogenesis (Chan et al., 2005), as well as a
stabilizing and transport moiety for extended action (Kim et al., 2010).

Table 1. Application of Transferrin in Fusion Proteins
Tf fused to:
Mouse-human chimeric IgG3
Targeted to the brain through Tf receptors on
the blood brain barrier (Shin et al., 1995)
Granulocyte colony-
stimulating factor (G-CSF)
Absorbed orally via Tf receptors in the GI tract
(Bai et al., 2005)
Vascular endothelial growth
factor (VEGF)
111In-labeled Tf enables imaging of VEGF
receptor-mediated angiogenesis in solid tumors
in vivo (Chan et al., 2005)
Glucagon-like peptide 1 (GLP-
1) or exendin-4 (EX-4)
Extended bioactivity (Kim et al., 2010)

25
CHAPTER II ．Plasmid Design and Cloning
I. ProINS-hTf Fusion Protein
Previous studies in the Shen lab constructed and characterized a human proinsulin-transferrin
(ProINS-hTf) fusion protein (Wang et al., 2011). This insulin mimetic exhibited a moderate and
sustained hypoglycemic effect in vivo (Wang et al., 2014). The fusion protein was also observed
to have liver specificity in terms of distribution and retention when compared to Tf and insulin
(Liu et al., 2018). ProINS-hTf fusion protein has a limited bioactivity of hypoglycemic control
unless it is converted to the active form, insulin-hTf (INS-hTf). The activation of this fusion protein
Figure 6. Plasmid map as indicated
AmpR refers to the ampicillin (Amp) resistance gene, allowing bacteria containing plasmids
growing in media with Amp. CMV promoter makes gene expressed in HEK293 cells. Enzyme
recognition sites are indicated as well. EcoRV, XhoI and XbaI are single-cut enzymes. SacI
cuts mTf rather than hTf. (Snapgene software, Insightful Science, available at snapgene.com)
)
26
is found to be mediated by Tf endocytosis and recycling pathway (Shao et al., 2016; Wang et al.,
2011). Besides the molecular size, the active INS-hTf protein can be retained in the liver through
a bivalent binding to both insulin receptor and Tf receptor on hepatocytes (Liu et al., 2020; Liu et
al., 2018), leading to a long-term action. The construction of ProINS-hTf fusion protein provides
a basis for the development of prohormone fusion protein engineering and potential protein therapy.
However, in long-term studies on ProINS-hTf, mice developed anti-human Tf antibody after a 2-
week daily injection of ProINS-hTf, prohibiting the studies to understand the long-term effects of
this fusion protein. To reduce immunogenicity of the fusion protein, the present study is designed
to exchange the Tf domain from human origin to mouse origin.

II. ProINS-mTf Plasmid Design
Previous studies tested the efficacy of ProINS-hTf in a NOD type 1 diabetic mouse model. The
mice developed antibodies against the ProINS-hTf within two weeks of injection. The lab
discovered that the detected antibody is anti-Tf rather than anti-insulin (data not shown). As human
proinsulin or insulin are small peptides, human Tf is more likely to be the cause of immune
response in mice. Therefore, to facilitate future experiments, the central hypothesis of this study
was that exchange of Tf from human origin to mouse origin will decrease the immune response in
mice. Based on the amino acid sequence (Fig.7), human Tf is ~88 % similar as mouse Tf and their
molecular sizes are similar to each other. Human Tf can specifically bind to the protein of
27

Figure 7. Pairwise
amino acid sequence
alignment between Tf
from human and mouse
(from EMBOSS)
The first row is human Tf
and the second is mouse.
The substitution matrix
is Eblosum62, gap
penalty 10.0 and extend
penatlty 0.5. One dot
indicates a small positive
score. Two dots indicate
a score > 1.0. The
alignment has length 702
with 72.8% matches,
1.6% matches.

28
meningococcal for iron uptake, whereas the mouse one does not (Zarantonelli et al., 2007).
Given that species-specific ligand has the highest affinity to the receptor, mouse Tf would bind
to its receptor at a higher affinity than human one (Penhallow et al., 1986) and is also likely to
lead to more suitable for experiments that need to be conducted long-term in mice.

The ProINS-hTf fusion protein was designed by cloning the proinsulin gene together with the
human Tf gene into pCDNA3.1(+) plasmid (Wang et al., 2011). In order to clone the mouse Tf
Figure 8. Study design
(A) Workflow of subcloning to replace hTf sequence with mTf sequence. The mTf
sequence is close to the hTf sequence regarding size. The reconstructed plasmid is also close
to the original one regarding size. (B) ProINS-Tf protein. Proinsulin and Tf are connected
though a LE dipeptide linker. The fusion protein included a His-tag to facilitate purification.

A
29
into this plasmid to replace the human Tf, the Tf Mouse Tagged ORF Clone plasmid (MR210104)
was obtained from Origene. This plasmid contains the mouse Tf gene (NM_133977) in the
backbone of pCMV6-Entry. To achieve the incorporation of the His-tag into the final fusion
product to facilitate protein purification, PCR cloning approach is used with mutagenic primers.
The mutagenic forward and reverse primers were designed as
5 -CCGCTCGAGGTCCCTGACAAAACGGTCAAATGGT-3 and
5-TGCTCTAGACTAATGATGATGATGATGATGGCTGCCCCCATGTTTGTGGA
AAGTGcaggcttcc-3.
The underlined basepairs encode the hexa His sequence, whereas the Gly-Gly-Ser sequence is
indicated in bold print. The primer design also contains a XhoI (5’-CTCGAG-3’) and a XbaI (5’-
TCTAGA-3’) restriction enzyme digestion sites, which allows subsequently ligation with the
ProINS-hTF plasmid through recognition sites of XhoI and XbaI.

III. Results
III.1 PCR to generate the insert
Subcloning process was designed to replace human Tf sequence with mouse Tf sequence. The
workflow is shown in Figure 8. Mouse Tf gene was amplified from mouse Tf plasmid through
PCR to generate the insert. During this process, His-tag, XhoI and XbaI recognition sites were
added to the end of the PCR product via designed primers. The PCR product is expected to be ~2
kb. The PCR product was applied onto the agarose gel and a major band of 2 kb was observed
30
(Fig.9A), indicating the expected PCR product. After double digestion of XhoI and XbaI, this 2
kb fragment containing mTf-GGSH6 sequence can be used as the insert for the resulting plasmid
in the ligation process (Fig.9B&9D).

III.2 Generating ProINS as receiving plasmid vector
The original plasmid encoding for ProINS-hTf fusion protein was digested by XhoI and XbaI
restriction enzymes, followed by the agarose gel separation and extraction for purification. After
digestion, the plasmid is expected to be cut into two fragments, which are ~6 kb and ~2 kb,
respectively (Fig.9C). The 6 kb fragment excluding hTf-GGSH6 sequence contains the ProINS
sequence and can be used as the receiving vector in the following ligation process. The digested
original plasmids were applied onto the agarose gel. Figure 9D shows a band of ~6 kb indicated
the vector.

III.3 Ligation
Ligation was performed with several different vector:insert ratios: 1:3, 1:5, 1:7. The ratio of 1:5
turned out to be the most optimum for this ligation as 7 colonies were obtained with the ligation
vs. 1 for ratio of 1:3 vs. 2 for ratio of 1:7 (data not shown). In the present study, the ligation was
performed using 1:5 (vector:insert). There were 11 colonies in total on the LB agar plates (plasmid
No.1 to No.11).
31
Plasmids were obtained using midi prep kit from each of these colonies. All plasmids were checked
for their sizes on DNA agarose gel first (Fig.10A: lane 1 to lane 11, plasmid No.1 to plasmid
No.11). For undigested plasmids running on the agarose gel, there are always more than one band.
Typically, there are four bands – nicked DNA, linear DNA, supercoiled DNA and circular, single-
Figure 9. Results of DNA agarose gel
(A) PCR product running on the gel. Lane 1 and lane 2 were from the same PCR
product. The band of 4 kb was produced by unspecific primer-binding. The band of 2 kb
was expected PCR product. (B) Purified PCR product digested by XhoI and XbaI.
Lane 1, 2, 3, and 4 were from the same digestion mixture. The band of 2 kb were expected
digested PCR product. (C) Double digestion of ProINS-hTf plasmid by XhoI and
XbaI. Lane 1, 2, 3, and 4 were from the same digestion mixture. The band of 2 kb
contained the hTf-GGSH6 sequence. The band of 6 kb was the vector, containing the
ProINS sequence. (D) Vector and insert check before ligation. The vector of ~ 6 kb was
loaded in lane 1. The insert of ~2 kb was loaded in lane 2.

32
stranded DNA from top to bottom in the lane on the gel. Though not all four bands can be observed
in a given plasmid preparation. The major band is generally the supercoiled DNA, which is of
smaller size than the linear DNA and mostly visible. The reconstructed plasmid is expected to be
~8 kb. In lanes 3, 7, 8, 9 and 10 (Fig.10A), a major band was below 6 kb and a minor band at
approximately 8 kb was observed above this major band. Therefore, these five plasmids (No.3, 7,
8, 9 and 10) are likely to be the expected ones, whereas other bands are unlikely to be the correct
plasmid.

During subcloning process, E. coli selection site in the plasmid was not changed. The sizes of
original (ProINS-hTf) and reconstructed plasmids (ProINS-mTf) are similar, which are 7808 bp
and 7805 bp, respectively. Therefore, using size difference based on their migration on agarose gel
cannot differentiate these two plasmids. By comparing the sequences of these two plasmids,
however, one site that is recognized by SacI restriction enzyme (5’-AGTACT-3’) is found in
mouse Tf sequence in ProINS-mTf and is absent from the human Tf sequence in ProINS-hTf
(Fig.6). Based on these findings, SacI restriction enzyme can be used to further examine the five
plasmids. In addition to this SacI site, two other SacI sites common to both ProINS-hTf and
ProINS-mTf are also present. Thus, if the plasmid contains mouse Tf sequence (ProINS-mTf),
three linear DNA fragments of ~1.6 kb, ~6.1 kb and ~100 bp will be expected after SacI digestion,
whereas there will be two bands of ~7.7 kb and ~100 bp if it is ProINS-hTf.

33
Figure 10B shows the results of the five selected plasmid plasmids (3, 7, 8, 9, 10) digested with
either SacI restriction enzyme or XbaI and XhoI restriction enzymes. XbaI and XhoI restriction
enzymes are the enzymes built into the plasmid design to separate the insert (mTf-GGSH6) and
the vector (ProINS), which are ~2 kb and ~6 kb, respectively. The digestion results show that
plasmid No.7 exhibited a band of ~1.6 kb (Fig.10B: lane 7-1) when incubated with SacI restriction
enzyme and two bands of ~6 kb and ~2 kb (Fig.10B: lane 7-2) when incubated with XhoI and XbaI
restriction enzymes. Plasmid No.8 showed a similar result as plasmid No.7. The digestion with
SacI, however does not appear to be complete. No.9 and No.10 showed the two expected bands in
digestion of XhoI and XbaI, and a band of ~8 kb in digestion of SacI. These two plasmids could
be ProINS-hTf plasmids. Due to incomplete digestion, a small amount of circular uncut ProINS-
hTf plasmids (~8 kb) may be extracted together with the vector in the band of 6 kb. E. coli
containing ProINS-hTf plasmids can also grow on the agar plate. No.3 showed one band of ~6 kb
in digestion of XhoI and XbaI, and one band of ~8 kb in digestion of SacI. While incomplete
digestion may have been the reason for not observing all the expected bands in these three colonies,
these results indicate that plasmid No.7 and plasmid No.8 are the most likely to be the expected
reconstructed plasmid.

To further verify that the sequence obtained is correct, plasmid No. 7 was sequenced and compared
with the mTf sequence in the Tf Mouse Tagged ORF Clone plasmid (MR210104) from Origene.
Several mutations were found in ProINS-mTf plasmid comparing to MR210104 (Table 2).
34

These mutations were found on nucleotide 1357, 1855, and 3172; and are G-A, A-C, C-T
respectively. These mutations, however, do not change the amino acid coding of the plasmid. In
addition, the PCR template, MR210104, was also sequenced. It appears that the mutation is
introduced prior to our PCR and ligation process as these mutations on plasmid No.7 were
identical to that of PCR template plasmid. Therefore, based on the results of enzyme digestion
and sequencing, it was confirmed that the human Tf sequence was successfully replaced with the
Figure 10. Plasmids extracted from colonies
(A) Uncut plasmids running on the agarose gel. Lanes 1 to 11 were uncut plasmids
extracted from different colonies. The lanes were numbered the same as plasmids. (B)
Plasmids (No. 3, 7, 8, 9, 10) cut by SacI or double digested by XhoI and XbaI. Lanes
a-b: “a” represents the label of the plasmid; “b” represents the number of enzymes added
in the reaction (“1” for SacI digestion, “2” for double digestion of XhoI and XbaI).

35
mouse Tf sequence in the resulting ProINS-mTf product. Plasmid No.7 is the correct
reconstructed plasmid.

Table 2. Sequence comparison and following translated amino acids among the record
from website, PCR templates and plasmid No.7
Location on
ProINS-mTf
plasmid
DNA sequence (5’ 3’)
MR210204 PCR templates (mouse Tf
plasmid) sequencing result
Plasmid No.7 (ProINS-mTf
plasmid) sequencing result
1357
G A A
AAG Lys AAA Lys
1855
A G G
CCA Pro CCG Pro
3172
C T T
TCC Ser TCT Ser

36
CHAPTER III. Expression and Characterization of ProINS-mTf Recombinant Fusion
Protein
To assess whether the resulting ProINS-mTf plasmid can indeed produce a fusion protein, the
newly constructed plasmid ProINS-mTf containing preproinsulin-mTf-GGSH6 sequence was
transiently transfected into HEK293 cells and HEK293T cells to express ProINS-mTf recombinant
fusion protein. Given that the protein machinery of HEK293 cells are applicable for non-
mammalian and mammalian proteins, HEK293 cells have been widely used for expression of
recombinant proteins (Thomas and Smart, 2005). In addition, HEK293 cells are easy to maintain,
feasible for various transfection methods, and highly efficient for protein production (Thomas and
Smart, 2005).

Previous studies concluded that ProINS-hTf fusion protein was efficiently secreted out of the cells
(Wang et al., 2011). As the vector in the newly constructed plasmid does not differ from that of
ProINS-hTf, ProINS-mTf fusion protein, if expressed, is expected to be found mostly outside the
cell after transcription and translation. To examine if the protein was expressed, cell media were
collected and concentrated using 4 mL Amicon® Ultra-4 Centrifugal Filter Units (C7719, Milipore
Sigma). The use of this filter allows for the concentration of proteins in the media from ~8 mL to
~100 L using a molecular size limiting membrane. The membrane used in this experiment has a
molecular size cutoff of above 50 kDa, which is expected to retain ProINS-mTf of ~90 kDa.
37

The concentrated cell culture media were then analyzed by Western blots to characterize the fusion
protein. After separating the protein, the membrane was probed with anti-human (Pro)INS
antibody. The result shows one major band with same molecular size of ~90 kDa in the transfected
Figure 11. Western blot result of concentrated cell media after transfection
A) Western blot. After concentration of conditioned media from HEK293T cells (Lanes 2-4)
or HEK293 cells (Lanes 5-7) protein samples were resolved in 10% SDS-PAGE, followed by
anti-human proinsulin Western blot. “M” refers to the protein marker (Lanes 1 and 8). “NC”
refers to the negative control (Lanes 2 and 5), in which transfection was not performed. “T1”
(Lanes 3 and 6) and “T2” (Lanes 4 and 7) represent different volumes of lipofectamine 3000
reagent added in each well, which were 3.75 ul and 7.5 ul, respectively. (B) Band density
(Lanes 3, 4, 6, and 7) quantified via ImageJ software. The density of band 3 is considered
A
B
38
cells (Fig.11A: lanes 3, 4, 6 and 7), whereas there were no bands in the non-transfected cells
(Fig.11A: lanes 2 and 5). This result indicates that the protein was expressed and secreted into the
media after transfection. ProINS-hTf is calculated to be ~89 kDa (Wang et al., 2011). The ProINS-
mTf fusion protein is expected to be approximately the same size. The band detected by the anti-
ProINS antibody in these transfected cells migrated at 90 kDa and thus is likely the ProINS-mTf
fusion protein. The use of HEK293 versus HEK293T cells did not appear to affect the protein yield
for this plasmid (Fig.11B). A low amount (3.75ul) of lipofectamine 3000 is sufficient for producing
the fusion protein.

39
CHAPTER IV. Discussion and Conclusion
I. Discussion
The present study is aimed to reconstruct a fusion protein by replacing human Tf with mouse Tf.
Our work started with mTf plasmid, MR210204, and is designed to clone the mTf piece from this
plasmid to the ProINS-hTf plasmid. The ProINS-hTf plasmid contains a human proinsulin gene
fused with Tf gene through a dipeptide linker, LE (Fig.6). The LE linker is the recognition site of
XhoI (5’-CTCGAG-3’). In addition, a Gly-Gly-Ser-hexa His sequence (-GGSHHHHHH- or -
GGSH6-) was incorporated into the carboxyl-terminal region of the full-length human Tf (residues
1–679) to make a His-tagged hTf (hTf-GGSH6) (Wang et al., 2011). This His-tag was incorporated
to facilitate purification of the fusion protein. His-tag can bind to nickel (Ni2+) at pH of 8 and
above. By changing condition in the nickel column, protein with a His-tag can bind on and off
Ni2+, and thereby the protein can be purified. Serving as a spacer, the tripeptide GGS makes His-
tag available for metal-binding (Wang et al., 2011). After plasmid expression, preproinsulin is first
synthesized. The N-terminal signal peptide of preproinsulin was removed before secretion in the
cell (Wang et al., 2011). In the subcloning process, the hTf-GGSH6 would be removed after
enzyme digestion. In order to maintain the GGSH6, PCR primers are designed to incorporate this
tag. With this design, the resulting plasmid is expected to produce a ProINS-mTf that has
comparable functions as the ProINS-hTf without inducing immunogenicity in mouse. In this thesis
work, ProINS-mTf plasmid was successfully constructed and transfected in HEK293 and
40
HEK292T cells using standard protein producing protocols. This work will allow future in vivo
work to test the efficacy of ProINS-Tf using a mouse model and advance its therapeutic
development for treatment of diabetes.

Several technical issues are encountered during the cloning of this plasmid. First, different strains
of E. coli are important at amplifying different plasmids. Using our laboratory-made competent
cells (DH5) that are verified to be competent E. coli with commercial purified plasmids and other
plasmids the lab has used, no colonies on the LB agar plates were obtained, even with un-
manipulated plasmids. To overcome this problem, the Top10 strain of E. coli was used. Per
suggestion by the manufacturer, the Top10 strain is likely to have a higher transformation
efficiency than DH5. This may be related to the enzymes produced by DH5 can methylate and
degrade foreign DNA. Top10 that has the same genotype as DH10 carries mutations in these
enzymes and is thus more tolerant to foreign DNA.

After switching to a different strains of E. coli (TOP10 from Invitrogen), colonies were
successfully obtained. These results indicate that the strains of competent E. coli display a higher
efficiency or tolerance of transformation. The storage time of our laboratory-made cells may
exceed 1 year, a typical shelf life of competent cells, thereby reducing the efficiency of competent
cells. As the ligation mixture contains much more amounts of non-ligated linear DNA fragments
41
and incorrectly ligated DNA than purified plasmids do. These fragments may also impact the
transformation efficiency of competent cells.

In this study, the size of colonies on the plate was smaller than expected even though they were
incubated for 24 hrs. This may be due to several reasons: 1) antibiotic concentration can be too
high to enable correct colonies to grow large on the plate; 2) competent cells are not viable enough
to grow large after transformation of the constructs; 3) incubation at 37 °C is not the most favorable
condition.

The yield of ProINS-mTf protein appears to be efficient regarding the result of Western blots. In
addition to the possibility that the ProINS-mTf is naturally productive, the type of transfection
reagent used may also account for the yield. The current study uses lipofectamine 3000 (Lipo3000)
whereas polyethylenimine (PEI) was used previously for ProINS-hTf. Lipo3000 are the same as
PEI method in terms of mechanism delivering plasmids into the cell, which is through the
formation of DNA-cationic lipid complex. In HeLa cells, Lipo2000 showed a significant higher
transfection efficiency than PEI (Lehner et al., 2013). Lipo3000 (protocol modified in 2016) is
improved based on Lipo2000 showing higher efficiency and protein expression, as well as lower
toxicity. In addition, a shorter time lapse was used for media collection (1 day vs. 3 days). This
may allow collection of products at the optimum expression time. Transient transfected plasmids
can be generally expressed for up to one week. The amount of protein production generally goes
42
up in the first a few days and reaches the maximum, followed by a reduction. Further investigation
on the amount and the purity of the fusion protein is needed to improve yield.

II. Conclusion
The present study reconstructed a fusion protein by replacing human Tf with mouse Tf. This work
will allow future in vivo work to test the efficacy of ProINS-Tf using a mouse model and advance
the therapeutic development of ProINS-hTf. Further investigation is needed to validate whether
the reconstructed ProINS-mTf protein can exhibit similar hypoglycemic effect as ProINS-hTf. In
addition, a major focus will be validating that the reconstructed protein does not elicit similar
immune reactions in the mice as ProINS-hTf. Overall, ProINS-Tf fusion protein has the potential
to be a new basal insulin therapy and the Tf receptor-mediated endocytosis and recycling pathway
can be a basis for the development of protein drugs.

43
Materials and Methods
Subcloning process. During XhoI and XbaI double digestion and agarose gel electrophoresis, hTf-
GGSH6 sequence, the band of 2 kb, was separated from the band of 6 kb. To retain the GGSH6
sequence in the reconstructed plasmid, PCR-based mutagenesis methods were used, during which
GGSH6 sequence can be added onto the full-length mouse Tf (residues 1-678). The PCR product
was purified by the agarose gel extraction, followed by double digestion of XhoI and XbaI. The
digested PCR product was loaded into the agarose gel for purification. The purified digested PCR
product, mTf-GGSH6 sequence, was referred to as the insert. Purified inserts and vectors were
incubated at 65°C for 10 mins to inactivate any active restriction enzymes before ligation. And
inserts and vectors (insert:vector = 5:1) were mixed before adding ligation buffer and T4 DNA
ligase. Ligation was mediated through T4 DNA ligase (from NEB). The ligation solution was
incubated at 4°C for 1 hour, followed by 16°C for 16 hours.

Cell culture. HEK293 and HEK293T cells were cultured in DMEM containing 10% FBS and 1%
Glutamine. DMEM medium was replaced with conditioned serum-free CD293 medium 6 hours
after transfection. Cells were cultured in a humidified incubator with 95% air and 5% CO2 at 37 °C.

Protein production. Poly-D-lysine was used to coat the 6-well plate first to avoid detachment of
HEK293 cells. Cells were plated onto 6-well plate before transfection. The plasmids encoding for
44
preproinsulin-mTf-GGSH6 fusion protein were transiently transfected to HEK293 cells and
HEK293T cells through lipofectamine 3000 reagent mediated DNA transfection. Conditioned
serum-free CD293 medium (Invitrogen) were collected four times every other day. Collected
media were concentrated using 4 mL Amicon® Ultra-4 Centrifugal Filter Units (C7719, Milipore
Sigma).

DNA agarose gel. All DNA samples were analyzed on 0.5-fold TAE (Tris-Acetate-EDTA) agarose
gel under 100 V.

Western Blot. Concentrated cell media containing same protein quantity (~10 ug) were loaded into
each well and separated on 10% SDS-PAGE gel. After transferring the proteins from the gel to the
membrane, anti-(Pro)INS antibody (GS-9A8, DSHB) was used to blot the protein on the
membrane.

Transformation. Ligated plasmid was transformed into One ShotTM TOP10 Chemically Competent
E. coli (C404010, Invitrogen) using heat-shock transformation.

Bacterial culture. All bacterial culture was performed at 37 °C with 250 rpm. LB (Luria-Bertani)
medium and LB agar plate were prepared as shown in the following table:

45
1L Tryptone Yeast Extract NaCl Agar
LB Medium 10 g 5 g 10 g 0
LB Agar Plate 10 g 5 g 10 g 15 g

DNA extraction. Plasmid extraction from E. coli was performed using miniprep and midiprep kits
(QIAprep spin kit, Qiagen). Gel extraction was performed using gel extraction kit (28704, Qiagen).

Polymerase chain reaction (PCR). PCR reaction mixture was prepared as below.
PCR
template
Forward or reverse
primers (10 uM)
DNA
polymerase
10 X PCR
reaction buffer
Total volume
600 ng 3 uL 0.8 uL 10 uL 100 uL
PCR reaction program was designed as below.
Cycle number Temperature 1 & Time 1 Temperature 2 & Time 2
1 95 °C 2 min X
30 95 °C 15 s 68 °C 2 min 30 s
1 68 °C 2 min X

Restriction enzyme digestion. All DNA restriction enzymes were from NEB. DNA samples were
incubated with enzymes at 37 °C for 4 hours or overnight.

Transfection. After cells were seeded onto 6-well plate, transfection of the reconstructed plasmids
was performed using LipofectamineTM 3000 transfection reagent (L3000008, ThermoFisher).
Protein assay. Concentrated cell media were quantified for protein using DCTM protein assay
(5000111, Bio-Rad).
46
Sequencing. All sequencing data were obtained through Genewiz Sanger Sequencing.
47
References
Arndt, T. (2001). Carbohydrate-deficient transferrin as a marker of chronic alcohol abuse: A critical
review of preanalysis, analysis, and interpretation. Clinical Chemistry 47, 13-27.
Association, A.D. (2004). Continuous subcutaneous insulin infusion. Diabetes Care 27, s110-s110.
Attri, A.K., Fernandez, C., and Minton, A.P. (2010). pH-dependent self-association of zinc-free
insulin characterized by concentration-gradient static light scattering. Biophys Chem 148, 28-33.
Awai, M., and Brown, E. (1963). Studies of the metabolism of I131-labeled human transferrin.
Journal of Laboratory and Clinical Medicine 61, 363-396.
Baeshen, N.A., Baeshen, M.N., Sheikh, A., Bora, R.S., Ahmed, M.M., Ramadan, H.A., Saini, K.S.,
and Redwan, E.M. (2014). Cell factories for insulin production. Microb Cell Fact 13, 141.
Bai, Y ., Ann, D.K., and Shen, W.-C. (2005). Recombinant granulocyte colony-stimulating factor-
transferrin fusion protein as an oral myelopoietic agent. Proceedings of the National Academy of
Sciences 102, 7292-7296.
Baker, E.N., Blundell, T.L., Cutfield, J.F., Cutfield, S.M., Dodson, E.J., Dodson, G.G., Hodgkin,
D.M., Hubbard, R.E., Isaacs, N.W., Reynolds, C.D., et al. (1988). The structure of 2Zn pig insulin
crystals at 1.5 A resolution. Philos Trans R Soc Lond B Biol Sci 319, 369-456.
Barnett, A.H. (2003). A review of basal insulins. Diabetic medicine 20, 873-885.
Becker, R., Wollmer, A., and Seipke, G. (2007). Self-association of insulin glulisine in solution -
Results of a comparative circular dichroism study. Diabetes 56, A124-A125.
Becker, R.H. (2007). Insulin glulisine complementing basal insulins: a review of structure and
activity. Diabetes Technol Ther 9, 109-121.
Becker, R.H., and Frick, A.D. (2008). Clinical pharmacokinetics and pharmacodynamics of insulin
glulisine. Clin Pharmacokinet 47, 7-20.
48
Bergenstal, R.M., Cohen, R.M., Lever, E., Polonsky, K., Jaspan, J., Blix, P.M., Revers, R., Olefsky,
J.M., Kolterman, O., Steiner, K., et al. (1984). The Metabolic Effects of Biosynthetic Human
Proinsulin in Individuals with Type I Diabetes*. The Journal of Clinical Endocrinology &
Metabolism 58, 973-979.
Bliss, M. (1993). The history of insulin. Diabetes Care 16, 4-7.
Bolli, G., Di Marchi, R., Park, G., Pramming, S., and Koivisto, V .A. (1999). Insulin analogues and
their potential in the management of diabetes mellitus. Diabetologia 42, 1151.
Bolli, G.B., and Owens, D.R. (2000). Insulin glargine. The Lancet 356, 443-445.
Brandsma, M.E., Jevnikar, A.M., and Ma, S. (2011). Recombinant human transferrin: beyond iron
binding and transport. Biotechnology advances 29, 230-238.
Brange, J., Owens, D.R., Kang, S., and Vølund, A. (1990). Monomeric Insulins and Their
Experimental and Clinical Implications. Diabetes Care 13, 923-954.
Brange, J., Ribel, U., Hansen, J., Dodson, G., Hansen, M., Havelund, S., Melberg, S., Norris, F.,
Norris, K., and Snel, L. (1988). Monomeric insulins obtained by protein engineering and their
medical implications. Nature 333, 679.
Brems, D.N., Alter, L.A., Beckage, M.J., Chance, R.E., DiMarchi, R.D., Green, L.K., Long, H.B.,
Pekar, A.H., Shields, J.E., and Frank, B.H. (1992). Altering the association properties of insulin by
amino acid replacement. Protein Engineering, Design and Selection 5, 527-533.
Castillo, J., Codina, M., Martínez, M.L., Navarro, I., and Gutiérrez, J. (2004). Metabolic and
mitogenic effects of IGF-I and insulin on muscle cells of rainbow trout. American Journal of
Physiology-Regulatory, Integrative and Comparative Physiology 286, R935-R941.
Chan, C., Sandhu, J., Guha, A., Scollard, D.A., Wang, J., Chen, P., Bai, K., Lee, L., and Reilly,
R.M. (2005). A human transferrin-vascular endothelial growth factor (hnTf-VEGF) fusion protein
containing an integrated binding site for 111In for imaging tumor angiogenesis. Journal of Nuclear
Medicine 46, 1745-1752.
49
Ciszak, E., Beals, J.M., Frank, B.H., Baker, J.C., Carter, N.D., and Smith, G.D. (1995). Role of C-
terminal B-chain residues in insulin assembly: the structure of hexameric LysB28ProB29-human
insulin. Structure 3, 615-622.
Dautry-Varsat, A., Ciechanover, A., and Lodish, H.F. (1983). pH and the recycling of transferrin
during receptor-mediated endocytosis. Proceedings of the National Academy of Sciences 80, 2258-
2262.
De Leeuw, I., Vague, P., Selam, J.L., Skeie, S., Lang, H., Draeger, E., and Elte, J. (2005). Insulin
detemir used in basal ‐bolus therapy in people with type 1 diabetes is associated with a lower risk
of nocturnal hypoglycaemia and less weight gain over 12 months in comparison to NPH insulin.
Diabetes, Obesity and Metabolism 7, 73-82.
Derewenda, U., Derewenda, Z., Dodson, E.J., Dodson, G.G., Reynolds, C.D., Smith, G.D., Sparks,
C., and Swenson, D. (1989). Phenol stabilizes more helix in a new symmetrical zinc insulin
hexamer. Nature 338, 594-596.
Dimitriadis, G.D., and Gerich, J.E. (1983). Importance of Timing of Preprandial Subcutaneous
Insulin Administration in the Management of Diabetes Mellitus. Diabetes Care 6, 374-377.
Emdin, S.O., Dodson, G.G., Cutfield, J.M., and Cutfield, S.M. (1980). Role of zinc in insulin
biosynthesis. Diabetologia 19, 174-182.
Fineberg, N.S., Fineberg, S.E., Anderson, J.H., Birkett, M.A., Gibson, R.G., and Hufferd, S. (1996).
Immunologic Effects of Insulin Lispro [Lys (B28), Pro (B29) Human Insulin] in IDDM and
NIDDM Patients Previously Treated With Insulin. Diabetes 45, 1750-1754.
Fineberg, S.E., Galloway, J.A., Fineberg, N.S., Rathbun, M.J., and Hufferd, S. (1983).
Immunogenicity of recombinant DNA human insulin. Diabetologia 25, 465-469.
Galloway, J.A., Hooper, S.A., Spradlin, C.T., Howey, D.C., Frank, B.H., Bowsher, R.R., and
Anderson, J.H. (1992). Biosynthetic Human Proinsulin: Review of Chemistry, in Vitro and in Vivo
Receptor Binding, Animal and Human Pharmacology Studies, and Clinical Trial Experience.
Diabetes Care 15, 666-692.
50
Glauber, H.S., Revers, R.R., Henry, R., Schmeiser, L., Wallace, P., Kolterman, O.G., Cohen, R.M.,
Rubenstein, A.H., Galloway, J.A., Frank, B.H., et al. (1986). In Vivo Deactivation of Proinsulin
Action on Glucose Disposal and Hepatic Glucose Production in Normal Man. Diabetes 35, 311-
317.
Haahr, H., and Heise, T. (2014). A review of the pharmacological properties of insulin degludec
and their clinical relevance. Clinical pharmacokinetics 53, 787-800.
Haak, T., Tiengo, A., Draeger, E., Suntum, M., and Waldhäusl, W. (2005). Lower within ‐subject
variability of fasting blood glucose and reduced weight gain with insulin detemir compared to
NPH insulin in patients with type 2 diabetes. Diabetes, Obesity and Metabolism 7, 56-64.
Hallschmid, M., Jauch-Chara, K., Korn, O., Mölle, M., Rasch, B., Born, J., Schultes, B., and Kern,
W. (2010). Euglycemic Infusion of Insulin Detemir Compared With Human Insulin Appears to
Increase Direct Current Brain Potential Response and Reduces Food Intake While Inducing
Similar Systemic Effects. Diabetes 59, 1101-1107.
Hansen, B.F., Danielsen, G.M., Drejer, K., Sørensen, A.R., Wiberg, F.C., Klein, H.H., and
Lundemose, A.G. (1996). Sustained signalling from the insulin receptor after stimulation with
insulin analogues exhibiting increased mitogenic potency. Biochemical Journal 315, 271-279.
Hansen, J.F. (1991). The self-association of zinc-free human insulin and insulin analogue B13-
glutamine. Biophys Chem 39, 107-110.
Havelund, S., Plum, A., Ribel, U., Jonassen, I., Vølund, A., Markussen, J., and Kurtzhals, P. (2004).
The mechanism of protraction of insulin detemir, a long-acting, acylated analog of human insulin.
Pharmaceutical research 21, 1498-1504.
Heise, T., Nosek, L., Spitzer, H., Heinemann, L., Niemoller, E., Frick, A.D., and Becker, R.H.
(2007). Insulin glulisine: a faster onset of action compared with insulin lispro. Diabetes Obes
Metab 9, 746-753.
Herring, R., Jones, R.H., and Russell-Jones, D.L. (2014). Hepatoselectivity and the evolution of
insulin. Diabetes, Obesity and Metabolism 16, 1-8.
51
Homko, C., Deluzio, A., Jimenez, C., Kolaczynski, J.W., and Boden, G. (2003). Comparison of
insulin aspart and lispro: pharmacokinetic and metabolic effects. Diabetes Care 26, 2027-2031.
Hordern, S., Wright, J., Umpleby, A., Shojaee-Moradie, F., Amiss, J., and Russell-Jones, D. (2005).
Comparison of the effects on glucose and lipid metabolism of equipotent doses of insulin detemir
and NPH insulin with a 16-h euglycaemic clamp. Diabetologia 48, 420-426.
Huang, X.F., and Arvan, P. (1994). Formation of the insulin-containing secretory granule core
occurs within immature beta-granules. Journal of Biological Chemistry 269, 20838-20844.
Ilag, L., Deeg, M., Costigan, T., Hollander, P., Blevins, T., Edelman, S., Konrad, R., Ortmann, R.,
Pollom, R., and Huster, W. (2016). Evaluation of immunogenicity of LY2963016 insulin glargine
compared with L antus® insulin glargine in patients with type 1 or type 2 diabetes mellitus.
Diabetes, Obesity and Metabolism 18, 159-168.
Jonassen, I., Havelund, S., Hoeg-Jensen, T., Steensgaard, D.B., Wahlund, P.O., and Ribel, U.
(2012). Design of the novel protraction mechanism of insulin degludec, an ultra-long-acting basal
insulin. Pharm Res 29, 2104-2114.
Kahn, C.R., and Rosenthal, A.S. (1979). Immunologic reactions to insulin: insulin allergy, insulin
resistance, and the autoimmune insulin syndrome. Diabetes care 2, 283-295.
Kalra, S. (2013). Insulin degludec: a significant advancement in ultralong-acting Basal insulin.
Diabetes Ther 4, 167-173.
Kalra, S., Mukherjee, J.J., Venkataraman, S., Bantwal, G., Shaikh, S., Saboo, B., Das, A.K., and
Ramachandran, A. (2013). Hypoglycemia: The neglected complication. Indian J Endocrinol Metab
17, 819-834.
Kim, B.-J., Zhou, J., Martin, B., Carlson, O.D., Maudsley, S., Greig, N.H., Mattson, M.P.,
Ladenheim, E.E., Wustner, J., Turner, A., et al. (2010). Transferrin Fusion Technology: A Novel
Approach to Prolonging Biological Half-Life of Insulinotropic Peptides. Journal of Pharmacology
and Experimental Therapeutics 334, 682-692.

52
Klausner, R.D., Ashwell, G., van Renswoude, J., Harford, J.B., and Bridges, K.R. (1983). Binding
of apotransferrin to K562 cells: explanation of the transferrin cycle. Proceedings of the National
Academy of Sciences 80, 2263-2266.
Kroeff, E.P., Owens, R.A., Campbell, E.L., Johnson, R.D., and Marks, H.I. (1989). Production
scale purification of biosynthetic human insulin by reversed-phase high-performance liquid
chromatography. J Chromatogr 461, 45-61.
Kurtzhals, P., Schäffer, L., Sørensen, A., Kristensen, C., Jonassen, I., Schmid, C., and Trüb, T.
(2000). Correlations of receptor binding and metabolic and mitogenic potencies of insulin analogs
designed for clinical use. Diabetes 49, 999-1005.
Ladisch, M.R., and Kohlmann, K.L. (1992). Recombinant human insulin. Biotechnology Progress
8, 469-478.
Lehner, R., Wang, X., and Hunziker, P. (2013). Plasmid linearization changes shape and efficiency
of transfection complexes. 5, 205.
Liu, Y ., Wang, H.-Y ., Shao, J., Zaro, J.L., and Shen, W.-C. (2020). Enhanced insulin receptor
interaction by a bifunctional insulin-transferrin fusion protein: an approach to overcome insulin
resistance. Scientific Reports 10, 7724.
Liu, Y ., Wang, H.-Y ., Zhou, L., Su, Y ., and Shen, W.-C. (2018). Biodistribution, activation, and
retention of proinsulin-transferrin fusion protein in the liver: Mechanism of liver-targeting as an
insulin prodrug. Journal of Controlled Release 275, 186-191.
Meneghini, L.F., and Miranda-Palma, B. (2012). Insulin degludec: a novel ultra-long-acting basal
insulin for use in Type 1 and 2 diabetes. Expert Review of Endocrinology & Metabolism 7, 9-14.
Penhallow, R.C., Brown-Mason, A., and Woodworth, R.C. (1986). Comparative studies of the
binding and growth-supportive ability of mammalian transferrins in human cells. J Cell Physiol
128, 251-260.
Pickup, J., and Keen, H. (2002). Continuous subcutaneous insulin infusion at 25 years: evidence
base for the expanding use of insulin pump therapy in type 1 diabetes. Diabetes care 25, 593-598.
53
Plank, J., Wutte, A., Brunner, G., Siebenhofer, A., Semlitsch, B., Sommer, R., Hirschberger, S.,
and Pieber, T.R. (2002). A direct comparison of insulin aspart and insulin lispro in patients with
type 1 diabetes. Diabetes Care 25, 2053-2057.
Quianzon, C.C., and Cheikh, I. (2012). History of insulin. J Community Hosp Intern Med Perspect
2, 10.3402/jchimp.v3402i3402.18701.
Radziuk, J.M., Davies, J.C., Pye, W.S., Shields, J.E., DiMarchi, R.D., and Chance, R.E. (1997).
Bioavailability and bioeffectiveness of subcutaneous human insulin and two of its analogs—
LysB28ProB29-human insulin and AspB10LysB28ProB29-human insulin—assessed in a
conscious pig model. Diabetes 46, 548-556.
Revers, R.R., Henry, R., Schmeiser, L., Kolterman, O., Cohen, R., Bergenstal, R., Polonsky, K.,
Jaspan, J., Rubenstein, A., Frank, B., et al. (1984). The Effects of Biosynthetic Human Proinsulin
on Carbohydrate Metabolism. Diabetes 33, 762-770.
Richards, J.P., Stickelmeyer, M.P., Flora, D.B., Chance, R.E., Frank, B.H., and DeFelippis, M.R.
(1998). Self-association properties of monomeric insulin analogs under formulation conditions.
Pharm Res 15, 1434-1441.
Rosenfeld, L. (2002). Insulin: Discovery and Controversy. Clinical Chemistry 48, 2270.
Schatz, H., and Ammermann, S. (1988). Human proinsulin: bioactivity and pharmaco-kinetics
after intravenous and subcutaneous administration. Horm Metab Res Suppl 18, 1-5.
Schernthaner, G. (1993). Immunogenicity and Allergenic Potential of Animal and Human Insulins.
Diabetes Care 16, 155-165.
Shao, J., Zaro, J.L., and Shen, W.-C. (2016). Proinsulin–Transferrin Fusion Protein Exhibits a
Prolonged and Selective Effect on the Control of Hepatic Glucose Production in an Experimental
Model of Type 1 Diabetes. Molecular pharmaceutics 13, 2641-2646.
Shin, S.U., Friden, P., Moran, M., Olson, T., Kang, Y .S., Pardridge, W.M., and Morrison, S.L.
(1995). Transferrin-antibody fusion proteins are effective in brain targeting. Proceedings of the
National Academy of Sciences 92, 2820-2824.
54
Shojaee-Moradie, F., Jackson, N.C., Boroujerdi, M., Brandenburg, D., Sönksen, P.H., and Jones,
R.H. (1995). Demonstration of a relatively hepatoselective effect of covalent insulin dimers on
glucose metabolism in dogs. Diabetologia 38, 1007-1013.
Slieker, L., Brooke, G., DiMarchi, R., Flora, D., Green, L., Hoffmann, J., Long, H., Fan, L., Shields,
J., and Sundell, K. (1997). Modifications in the B10 and B26–30 regions of the B chain of human
insulin alter affinity for the human IGF-I receptor more than for the insulin receptor. Diabetologia
40, S54-S61.
Stammberger, I., Seipke, G., and Bartels, T. (2006). Insulin glulisine - A comprehensive preclinical
evaluation. Int J Toxicol 25, 25-33.
Starr, J.I., and Rubenstein, A.H. (1974). Metabolism of Endogenous Proinsulin and Insulin in
Man*. The Journal of Clinical Endocrinology & Metabolism 38, 305-308.
Stein, B., and Sussman, H. (1986). Demonstration of two distinct transferrin receptor recycling
pathways and transferrin-independent receptor internalization in K562 cells. Journal of Biological
Chemistry 261, 10319-10331.
Stibler, H. (1991). Carbohydrate-deficient transferrin in serum: a new marker of potentially
harmful alcohol consumption reviewed. Clin Chem 37, 2029-2037.
Stoorvogel, W., Geuze, H.J., Griffith, J.M., and Strous, G. (1988). The pathways of endocytosed
transferrin and secretory protein are connected in the trans-Golgi reticulum. The Journal of Cell
Biology 106, 1821-1829.
Thomas, P., and Smart, T.G. (2005). HEK293 cell line: a vehicle for the expression of recombinant
proteins. J Pharmacol Toxicol Methods 51, 187-200.
Tibaldi, J. (2007). Actions of insulin beyond glycemic control: a perspective on insulin detemir.
Advances in therapy 24, 868-882.
Tibaldi, J.M. (2014). Evolution of Insulin: From Human to Analog. The American Journal of
Medicine 127, S25-S38.
55
Tottrup, H.V ., and Carlsen, S. (1990). A process for the production of human proinsulin in
Saccharomyces cerevisiae. Biotechnol Bioeng 35, 339-348.
Tsuchiya, Y ., Saito, M., Kadokura, H., Miyazaki, J.-i., Tashiro, F., Imagawa, Y ., Iwawaki, T., and
Kohno, K. (2018). IRE1–XBP1 pathway regulates oxidative proinsulin folding in pancreatic β
cells. The Journal of Cell Biology 217, 1287-1301.
van Bon, A.C., Bode, B.W., Sert-Langeron, C., DeVries, J.H., and Charpentier, G. (2011). Insulin
glulisine compared to insulin aspart and to insulin lispro administered by continuous subcutaneous
insulin infusion in patients with type 1 diabetes: a randomized controlled trial. Diabetes technology
& therapeutics 13, 607-614.
Wang, F., Surh, J., and Kaur, M. (2012). Insulin degludec as an ultralong-acting basal insulin once
a day: a systematic review. Diabetes Metab Syndr Obes 5, 191-204.
Wang, Y ., Chen, Y .-S., Zaro, J.L., and Shen, W.-C. (2011). Receptor-mediated activation of a
proinsulin-transferrin fusion protein in hepatoma cells. Journal of Controlled Release 155, 386-
392.
Wang, Y ., Shao, J., Zaro, J.L., and Shen, W.-C. (2014). Proinsulin-transferrin fusion protein as a
novel long-acting insulin analog for the inhibition of hepatic glucose production. Diabetes 63,
1779-1788.
Weiss, M.A. (2009). Proinsulin and the genetics of diabetes mellitus. Journal of Biological
Chemistry 284, 19159-19163.
Yosten, G.L., Maric-Bilkan, C., Luppi, P., and Wahren, J. (2014). Physiological effects and
therapeutic potential of proinsulin C-peptide. American Journal of Physiology-Endocrinology and
Metabolism 307, E955-E968.
Zarantonelli, M.L., Szatanik, M., Giorgini, D., Hong, E., Huerre, M., Guillou, F., Alonso, J.M.,
and Taha, M.K. (2007). Transgenic mice expressing human transferrin as a model for
meningococcal infection. Infect Immun 75, 5609-5614.

56
Zwickl, C., Smith, H., Zimmermann, J., and Wierda, D. (1995). Immunogenicity of biosynthetic
human LysPro insulin compared to native-sequence human and purified porcine insulins in rhesus
monkeys immunized over a 6-week period. Arzneimittel-Forschung 45, 524-528.

Abstract (if available)

Abstract Insulin mimetics has been developed since the discovery of insulin in 1921. The ultimate goal is to normalize blood glucose level in patients with diabetes. Previous study has designed and characterized a human proinsulin-transferrin (ProINS-Tf) recombinant fusion protein, which is a long-term insulin prodrug with liver selectivity. However, it showed immune reaction counteracting the effect of the human fusion protein in mouse models, prohibiting further in vivo studies. To overcome this immunogenicity, the present study is designed to replace human transferrin with mouse transferrin through subcloning. The resultant plasmid was successfully reconstructed and the protein produced in HEK293 cells was identified by immunoblotting. This thesis work will allow further studies to validate ProINS-Tf as a potential therapy for treating diabetes in patients.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue

PDF

Mechanism study of proinsulin-transferrin fusion protein as a long-acting and liver-targeting insulin prodrug

PDF

Proinsulin-transferrin recombinant fusion protein: mechanism of activation and potential application in diabetes treatment

PDF

Proinsulin-transferrin fusion protein as a liver targeting insulin therapeutic in Type 1 diabetes treatment

PDF

Human growth hormone-transferrin recombinant fusion protein for oral delivery

PDF

Preparation and characterization of Tf-G-CSF fusion protein

PDF

Characterization of actin based motility in mammalian cells through LIM and SH3 domain protein 1 (LASP1) and elastin like polypeptide (ELP) fusion protein

PDF

Tubulin-based fusion proteins as multifunctional tools

PDF

Comparison of the pharmacokinetic profiles between monomeric and oligomeric transferrin

PDF

Cognate receptor fusion proteins enable parenteral delivery of challenging rapalogues

PDF

Using cyclotides as a bioscaffold to target intracellular protein-protein interactions

PDF

Downregulation of osteopontin by PTEN in liver-specific Pten-null mouse model

PDF

Characterization of the transferrin oligomer and its potential application in drug delivery

PDF

Three advancements in biotechnology: new tools for synthetic biology and next generation sequencing

PDF

Developing recombinant single chain Fc-dimer fusion proteins for improved protein drug delivery

PDF

Sustained and targeted delivery of rapamycin using FKBP-elastin like polypeptide fusion proteins

PDF

Downregulation of METTL7A gene in liver-specific Pten-null mouse model

PDF

Inclusion body purification of elastin-like-polypeptide fusion proteins with a low transition temperature

PDF

Comparing drug release kinetics of rapamycin bound FKBP-ELP fusion proteins

PDF

Cellular uptake mechanism of elastin-like polypeptide fusion proteins

Asset Metadata

Creator Huang, Yijing (author)

Core Title Reconstruction of a proinsulin-transferrin fusion protein

School School of Pharmacy

Degree Master of Science

Degree Program Pharmaceutical Sciences

Publication Date 07/23/2020

Defense Date 06/30/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Diabetes,fusion protein,insulin,insulin mimetics,OAI-PMH Harvest,proinsulin,recombinant DNA,transferrin

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Stiles, Bangyan (committee chair), Okamoto, Curtis (committee member), Zaro, Jennica (committee member)

Creator Email yijingh2020@gmail.com,yijinghu@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-341089

Unique identifier UC11665393

Identifier etd-HuangYijin-8735.pdf (filename),usctheses-c89-341089 (legacy record id)

Legacy Identifier etd-HuangYijin-8735.pdf

Dmrecord 341089

Document Type Thesis

Rights Huang, Yijing

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA