University of Southern California Dissertations and Theses

Role of CD33 structure and function in Alzheimer’s disease

(USC Thesis Other)

Role of CD33 structure and function in Alzheimer’s disease

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Role of CD33 structure and function in
Alzheimer’s Disease

By

Mingke Wu
1

Mentor: Tobias S. Ulmer
1,2

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfilment of the Requirements for the Degree
MASTER OF SCIENCE
(Biochemistry and Molecular Medicine)
August 2019

2
Acknowledgements

First of all, I want to express my sincere gratitude to my mentor, Dr. Tobias S.
Ulmer for allowing me join his lab and the full support to my master project.
Without his help, I can't imagine how I could transfer from Analytical Chemistry to
Biochemistry so smoothly. He never blames me for the mistake I make, instead of
that, he tells me how to fix it and how to prevent it happen again. Every minute I
spent in his lab was both valuable and joyful. As everyone knows, being a scientist
means you have to tolerate hundreds times failure, which really makes you
depression and hopeless. But Dr. Ulmer wholeheartedly considerate to students,
using his enthusiasm of protein and patience light my hope every time. Thanks him
for being such an excellent mentor.

Secondly, I want to thank Alan Situ, my lab mate, who has taught me a lot of
experiment skills and trained me from a freshman to a skilled researcher. Besides,
he also helped me for some data acquiring task, including showing me some
website function and lab equipment’s instruction. When I got stuck, he always
analyzed the weird result I met and comforted me.

Next, I want to extend my gratitude to Dr. Ansgar B Siemer and Dr. Ralf Langen.
Dr. Ansgar B Siemer pointed out some problem I might meet in the future study at
the first committee meeting, which corrected my direction of research. Dr. Ralf
Langen told me how to arrange my thesis structure and reminded me of taking care
of the research’s timeline. During my first period of research, I do meet a lot of
problem. Both of them answered my question professionally and patiently.

3

Also, I would like to thank Monica Pan, the advisor of our program. She is so
considerate from the first time I met her. Providing useful information timely,
replying email immediately and keeping contact almost all the time. She also cared
students’ mental health condition, trying to use some warm words to ease our
nervous.

Last but not the least, I want to thank my family. They gave birth to me and let me
have the opportunity to study aboard. They supported me at the financial aspect and
made me have a harbor to stay whenever I feel tired. My whole family including
my parents, grandparents, aunts, uncles and cousins accompanied me through every
ups and downs during my life abroad.

4
Table of contents
Acknowledgements ………………………………………………………… 2
Table of contents …………………………………………………………… 4
Abstract …………………………………………………………………… 5
1. Introduction
1.1 Dementia …………………………………………………………… 7
1.2 Alzheimer’s Disease ………………………………………………… 8
1.3 CD33’s Function in cell signaling …………………………………… 10
1.4 Heteronuclear NMR Spectroscopy …………………………………… 13
2. Materials and Methods
2.1 Materials …………………………………………………………… 15
2.1.1 pET-44 plasmid (Novagen, Inc.) ………………………………… 15
2.1.2 BL-21(DE3) competent E. Coli Cells …………………………… 15
2.1.3 DNA Expression & Sequence Analyzation ……………………… 15
2.2 Methods ……………………………………………………………… 17
2.2.1 Protein expression …………………………………………………17
2.2.2 Protein purification ……………………………………………… 18
2.2.3 SDS phage analyzing ………………………………………… 21
2.2.4 NMR sample Preparing ……………………………………… 22
3. Results
3.1 Purification of CD33(K283-Q364) by FPLC Chromatography ……… 23
3.2 Cleavage of MBP-CD33(KQ) by TEV protease ……………………… 24
3.3 NMR spectroscopy of CD33(Q233-Q364), GGH-CD33(V257-R291) and
COMMD3 …………………………………………………………… 27
4. Discussion & Conclusion
4.1 Choice of CD33 fusion protein ……………………………………… 31
4.2 Degradation of the CD33 cytosolic domain by TEV protease ………… 31
4.3 Future directions ……………………………………………………… 33
5. References …………………………………………………………… 34
5
Abstract

The molecular origins of Alzheimer’s Disease (AD) onset and progression are still
incompletely revealed. A major accepted cause of AD is the aggregation of fibrotic amyloid
peptides (Aβ) in the human brain. We do not have any efficient therapy to treat AD for now .
Genetic studies show that Cluster of differentiation 33 (CD33) is a risk factor for AD
[1]
. This
transmembrane receptor binds a sialic acid binding site and resembles an immunoglobulin-
like lectin. CD33 can regulate innate immunity, but its functions in the brain still remain
unknown.

Alzheimer’s Disease is one of the most prevalent chronic neurodegenerative disease, the
most common cause of dementia and the sixth leading cause of death in the United States.
From the world report of Alzheimer’s disease 2018, there is 50 million people worldwide
living with dementia. By using some statistic model, the evaluation of people who living
with dementia will more than 152 million by 2050
[2]
. CD33 is located in the plasma
membrane of microglia membrane and can influence microglia phagocytosis of Aβ, which
alleviates Alzheimer’s Disease
[3]
.

Multidimensional heteronuclear NMR spectroscopy is a major technique for determining
the structure of bio macromolecules in solution. In my thesis, I started to examine the
CD33 structure by multidimensional, heteronuclear NMR spectroscopy. Specifically, the
transmembrane domain and intercellular domain, which includes a SHP (Src Homology
Phosphatase) binding site, were examined. Also we will use Paramagnetic-Tag to label TM
domain to study the aggregation state of the TM domains, which may contribute to cell
signaling. After finishing the structure work, we will move to use ITC (Isothermal Titrate
Calorimetry) to characterize the binding affinity between the cytosolic domain of CD33
6
and some candidate protein. Our work aims at finding some molecules which have the
potential to accelerate the microglia phagocytosis to treat AD.

7
1. Introduction

1.1 Dementia
Dementia is a common geriatric syndrome, an umbrella term for describing symptom
including cognitive decline, memory loss and gradually reducing capacity for handling
daily physical activities. Based on demographics of dementia
[4]
, the 21th century will be
the population aging and fast growth of the elderly population. Alzheimer’s Disease
International (ADI), as a reliable source, highlighted the fact that the number of persons
living with dementia increased rapidly. There was a survey from World Health
Organization(WHO) aimed at studying the prevalence of dementia, showing there was
about 47.5 million people living with dementia worldwide at 2016. They also made a
projection that this number might triple by 2050.

Dementia has been considered as unpreventable and untreatable for a long time. But we do
make a lot of progress on studying this type of disease. There are 11 types of dementia
including Creutzfeldt-Jakob Disease, Lewy Body Dementia, Down Syndrome and AD,
Frontotemporal Dementia, Posterior Cortical Atrophy, Parkinson’s Disease, V ascular
Dementia and Korsakoff Syndrome.
[5]

The Central Nervous System(CNS) plays an important role in neurodegeneration diseases.
CNS consists of the brain and spinal cord. Brain is the most complex organ in our body,
which is also directly related to dementia. That tissue consists of a hundred billion neurons
or so and each neuron connect to thousands more. There are many different immune
responses observed at various stages of neurodegeneration diseases.
[6]
Many kinds of these
responses may exacerbate the disease, but, on the other hand, can also be potential
therapeutic targets.

8
1.2 Alzheimer’s Disease
Alzheimer’s disease (AD) is a severe chronic neurodegenerative disease, the symptoms of
which include memory loss, cognitive impairment and even personality change. A statistic
research shows that around 60 percent of dementia cases are due to AD and it affects 12%
of the aged population in the world.
[7]
The etiology of AD onset and progression are still
incompletely revealed. A major accepted cause of AD is aggregation of fibrotic amyloid
peptides (Aβ) in human brain. Aβ peptide is generated in the brain and other organs by
cleaving the amyloid precursor protein(APP) by b- and g-secretases. In contrast, Aβ’s
concentration will keep increasing in patient’s brain for their defective clearing ability. That
high concentration Aβ later will undergo self-aggregation and assemble into entities
ranging from oligomers to protofibrils, fibrils and amyloid plaques(Fig.1).
There are two types of AD, one is early-onset Alzheimer’s Disease(EOAD), which
represents less than 5% of AD, and another one is late-onset AD(LOAD), accounting for
more than 95% of cases. Based on family studies, it is estimated that genetic factors play a
role in approximately 80% of AD disease.
[9]
EOAD with onset younger than 65 years of
age shows significant different from LOAD.
For EOAD, patients have greater parietal atrophy, more white matter abnormalities, and
less hippocampal volume loss, compared to those with LOAD.
[10]
Besides, mutations
found in EOAD including APP , PSEN1 and PSEN2 have been well characterized. APP is
β-amyloid precursor protein, identified as a causative gene of AD. Several mutations in
APP gene will result in the overproduction and aggregation of Aβ42. PSEN1 and PSEN2
were identified in family linkage studies in 1995. Both genes products constitute the
enzymatic centre of the g-secretase complex, which response for amyloidogenic pathway
of APP cleavage. The PSEN1 mutations account for around 80% cases of EOAD and the
PSEN2 mutations only account for approximately 5%.
[11]
9
Figure 1. APP metabolism, oligomerization and signaling involved in neuronal
damage in AD.
Proteolytic cleavage of APP by β- and γ-secretase results in the generation of the Aβ monomers, which
under pathological conditions can assemble into potentially toxic oligomers. Oligomers can be sequestered
into fibrillary aggregates in plaques and for the toxic Aβ species that contribute to dysregulation of signaling
pathways and result in alterations to cytoskeletal and synaptic proteins and subsequent synaptic and
neuronal damage. Aβ accumulation is mediated by factors including rates of peptide production,
aggregation and clearance. Figure taken from ref
[8]
.

LOAD, in contrast, is more sporadic AD. Single nucleotide polymorphisms(SNPs) in a
number of genes have an effect on LOAD, which also combined with environmental
factors. APOE, as the main cholesterol transporter in the brain, by binding to Aβ, may play
an important role in AD pathogenesis. There is a strong genetic risk factor called APOE-
E4, whose heterozygous carriers risk for AD is increased three times and homozygous
carriers increasing 10-12-fold.
[12]
Genome-wide association studies (GW AS) also reveal
numerous risk genetic loci expressed at high level in microglia associated with AD,
including ABCA7, CR1, MS4A6A, TREM2, CD33 and INPP5D.
Immune response has been demonstrated to be part of pathogenesis of AD. Inflammation
in central nervous system(CNS) might cause the immune system dysfunction. Microglia,
the innate immune cells of the CNS, develop from erythromyeloid progenitor cells and
10
later on migrate into brain tissue. The main functions of microglia are immune defense and
maintaining homeostasis of CNS.
[13]
In response to changes in micro-environment and cell
signaling, microglia will exhibit multidimensional and environment-dependent phenotypes
that can be either cytotoxic or neuroprotective.

1.3 CD33’s function in cell signaling

CD33 is a transmembrane receptor belonging to the family of sialic acid-binding
immunoglobulin-like lectin (Siglecs). They are expressed highly on haematopoetic cell
lineages, compared to their expression level in differentiated cells. The main functions of
CD33 are participating in the discrimination between self and non-self, and recognizing the
glycan ligands to regulate the function of cells in the innate and adaptive immune system.
For their structure, the extracellular
domain has an Ig-like V-type and C2-type
domain, the V type domain is the
sialylated ligand binding site (Figure 2).
The cytoplasmic domain has two tyrosine
residues (Y340 & Y358), which function
as one of the most important downstream
regulation part. When they are
phosphorylated, these two residues could
accept phosphatase SHP-1 and/or SHP-2,
enabling CD33 to be an inhibit
downstream signaling.
[14]

Figure 2. CD33 structure. The left one is
wild type human CD33, the right one is
mutant Ig-V type domain.
11

CD33 has a membrane proximal immunoreceptor tyrosine based inhibitory motif (ITIM)
and a membrane-distal ITIM like motif.
[15]
ITIM motifs serve as docking sites for
phosphatase and regulate phosphatase-related inhibitory effects. ITIMs in cytoplasmic tail
have a consensus amino acid sequence, namely (I/V/L/S)-X-Y -X-X-(L/V), where X
denotes any amino acid. Src family PTKs (protein tyrosine kinase) phosphorylate the
tyrosine residue in ITIM. These phosphotyrosine can later on recruit tyrosine phosphatase
(SHIP , SHP1 and SHP2), which can directly remove phosphate groups from various
substrate proteins and lead to a block in early PTK activation, thus playing a major role in
balancing cellular response or controlling over-activation by antagonizing with the signals
coming from immunereceptor tyrosine-based activation motifs (IT AMs).
[16]
Figure 3
shows the CD33 signaling pathway of with and without sialic acids-siglec interaction and
how they regulating the phosphatase-mediated deactivation.
[17]

Originally, CD33 was studied as therapeutic target for Acute Myeloid L eukemia(AML).
Leukemia results from transformed multipotent CD33
-
stem cells or committed CD33
+

myeloid precursors. Using AMG 330 in primary AML sample ex vivo led to T cell
recruitment and expansion, this remarkable antibody can mediate cytotoxicity. Targeting
the CD33 differentiation antigen in vivo with a promises to be an effective therapeutic
approach.
[19]

12

In AD, the main role of CD33 is to inhibit microglia to phagocytose extracellular Aβ.
(Figure 4). Thus, CD33 inactivation improves phagocytosis and mitigates Aβ pathology.
Based on that principle, CD33 inhibition may become a novel therapy of AD and our
work, understanding the structural basis of CD33 signaling, is to contribute to achieve this
goal.
Figure 3. Schematic Diagram of the proposed pathway of CD33 siglec-mediated
suppression of cellular function. On macrophage, after ligand binding with CD33 related siglecs,
ITIM present on the cytosolic portion of siglecs get activated by phosphorylation with SH2 family
kinases (Lyn, Syk). Then Src homology 2 domain-containing phosphatases 1 and 2 (SHP1 and SHP2)
bind with the ITIM and get activated. These activated SHPs further dephosphorylate various signaling
molecules and suppress cellular activation. The target of SHPs might be p38 mitogen-activated protein
kinases (p38 MAPK) or Akt or nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kβ)
and ultimately downregulate the effector functions of the cell. In macrophages this siglec-mediated
pathway might regulate the polarization of macrophage function towards anti-inflammatory type by
reducing Th1 cytokines (IFNγ) and increasing Th2 cytokines (IL-10) along with downregulating
inducible nitric oxide synthase (iNOS) gene expression for nitric oxide (NO) secretion.

13

Figure 4. CD33 regulates microglial phagocytosis. Wild-type microglia express a low level of
CD33. The AD risk allele rs3865444 is associated with higher levels of CD33 that suppress the microglial
phagocytosis. Conversely, knockdown of CD33 significantly improves the phagocytosis capability of
microglia
1.4 Heteronuclear NMR Spectroscopy

NMR spectroscopy is an elegant method to study biological macromolecules’
structure, interactions and dynamics properties. It combines multiple scientific
fields including physics, chemistry, biology and medicine.
Specifically, this technique provides information about atomic coordinates and
dynamic motions of protein in solution or, in case of membrane proteins, when
solubilized in lipids or detergents.
[18]

NMR spectroscopy excites atomic nuclei with electromagnetic waves and observes
their mutual interactions in space and through covalent backbones. Since every
nucleus can be observed separately, the interactions between nuclei can be
resolved for proteins up to approximately 40 kDa. For proteins, NMR can
determine the three-dimensional structure. Different amino residues have different
side chains which determining they show unique chemical shift. In addition, the
chemical shifts of amino acids are also affected by their structural environment,
14
which is determined by the three-dimensional fold of the protein. Nonehtless, the
larger the protein, the more overlap of signal is obtained and isotope labeling can
alleviate this limit considerably (Figure 5).

Figure 5. Resonance Assignment Strategies Depend on the Protein’s MW . As the
molecular weight increase the number of peaks also increase, resulting in crowded and overlapped spectra.
Additionally, the proteins tumble slower in solution which result in broader peaks.

15

2. Material and Methods
2.1 Materials
2.1.1 pET-44 plasmid (Novagen, Inc.)
pET-44 plasmid is a powerful system for the cloning and high-level expression of
recombinant polypeptide fused with the 495 AA NusA (Nus.Tag) protein in E. coli.
pET-44, which driven by the bacteriophage T7 promoter, encode an additional N-terminal
His-Tag sequence for enhanced yields, versatile detection and purification of target
proteins. It also has ampicillin antibiotic resistance.

2.1.2 BL-21(DE3) competent E. Coli Cells
BL21(DE3) pLysS-T1
R
competent Escherichia Coli is an expression stain design for high
level induction and expression of genes from T7 promoter-based expression vector. DE3 is
designed for lysogenic for a lambda prophage containing an inducible T7 RNA
polymerase. The pLysS plasmid expresses lower levels of T7 lysozyme, which function as
inhibitor of the T7 RNA polymerase. T7 lysozyme lowers the background expression level
of target genes under the control of the T7 promoter but does not interfere with the level of
expression achieved following induction by IPTG.

2.1.3 DNA Expression & Sequence Analyzation
CD33 Full Length(Dharmacon)
The full length of CD33 construct is as follows,
16

The extracellular domain region is marked by dark blue, the transmembrane region is
marked by ginger yellow, and the red portion is the cytosolic domain.

Chart 1: CD33 sequences’ parameters

Name of Peptide Number of
Amino
Acids
Molecular
Weight/ Daltons
Theoretical
PI
Extinction
Coefficient(280nm)
GB3R2-TEV-
CD33(Y233-Q364)
214 22778 7.23 15930
CD33(Y233-Q364)
133 14034 8.87 4470
GB3R2-TEV-
CD33(Q251-R291)
122 13119 8.59 11460
CD33(Q251-R291)
42 4376 10.86
149580(at 205nm)
GB3R2-TEV-GGH-
CD33(V257-R291)
118 12671 8.59 11460
GGH-CD33(V257-R291)
38 3927.78 10.93
141910(at 205nm)
GB3R2-TEV-
CD33(Q251-Q364)
195 20961 7.23 14440
CD33(Q251-Q364)
117 12460 8.55 2980
MBP-TEV-CD33 (K283-
Q364)
490 53803 6.27 70820
CD33 (K283-Q364)
83 9021 9.25 2980
17
MBP-TEV-CD33 (G295-
Q364)
477 52369 5.90 70820
(G295-Q364)
70 7588 6.19 2980

2.2 Methods
2.2.1 Protein expression
The plasmid pET44-GB3R2-TEV-CD33 and pET44-MBP-TEV-CD33 described above
are transformed into the E. coli strain BL21(DE3) pLysS-T1R and grown on LB plates
with ampicillin (100 µg/mL) and chloramphenicol (50 µg/mL). For MBP-TEV-CD33, we
use LB plates with ampicillin (100 µg/mL) only.
Single colonies are picked to inoculate a 2mL LB Lennox culture containing 50 µg/mL
ampicillin and 34 µg/mL Chloramphenicol and grown at 37 °C and 200 rpm for 6-8 h.
This starting culture is then used to inoculate another 75mL starting culture, which is
grown overnight at 37 °C and 200 rpm. After centrifugation, the cells are transferred to a 1
L expression culture and grown at 37 °C and 200 rpm. At OD
600
=0.8, protein expression is
induced with IPTG at a final concentration of 0.5 mM for 4 hours at 37°C.
All culture media formulas are shown in Table 2. IPTG (Isopropyl ß-D-1-
thiogalactopyranoside) functions as a molecular mimic of allolactose, a lactose metabolite
that triggers transcription of the lac operon, inducing protein expression of genes under the
control of the lac operator.

Chart 2: Growth Culture Formula
20X Minimal Media Amount per liter
Na
2
HPO
4
·7H
2
O 120g
KH
2
PO
4
60g
18
NaCl 10g
15
N-Culture Media 75mL 1L
15
NH3·Cl 0.075g 0.5g
20X MM 3.75mL 50mL
H
2
O 70mL 930mL
0.1 M CaCl
2
75µL 1mL
1 M MgSO
4
75µL 1mL
Chloramphenicol (34mg/mL) 75µL 1mL
Ampicillin (1g/10mL) 37.5µL 500µL
Thiamine(20mg/mL) 37.5µL 500µL
20% Glucose 1.5mL 20mL
Media sterilization (All types): Autoclave liquid for around 80 min.
Chemicals labeled in red are added before culture use.

2.2.2 Protein purification
Immobilized metal ion affinity chromatography (IMAC) using Ni(II)SO
4
selectively
retains His-tagged proteins whereas untagged proteins pass through the column.
Specifically, I used the AKT A prime plus FPLC system (Amersham Bioscience) and 5 mL
GE HiTrap IMAC HP column (cat no. 17-0409-03) to carry out IMAC.

Cells were harvested by centrifugation at 4 °C and 4000g for 20min. (SOR V ALL SLC-
6000 ROTOR), and re-suspend in lysis buffer (50 mM Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300
mM NaCl, 20 mM imidazole, 100 mM SDS, pH readjusts to 8.0 after add imidazole).
Cells were lysed by sonication (MISONIX Sonicator 3000) until no longer viscous.

19
1
st
Purification:
The GE HiTrap IMAC HP columns are stored in 20% Ethanol at 4 °C. To remove the
ethanol in column to prevent the salt we use later precipitating, first we use 6 CV MQ H
2
O
wash at 5mL/min. Charge the 5 mL GE HiTrap IMAC HP column with 5mL of 0.1 M
NiSO
4
at 2mL/min. Then rinse excess NiSO
4
with 6 CV MQ H
2
O at 2 mL/min.
Use 30mL wash buffer I (50 mM Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300 mM NaCl, 25 mM
SDS, 2mM β-mercaptoethanol) to equilibrate the column at 5mL/min.
Load the supernatant of the sample in lysis buffer at 1 mL/min. W ash with approximately
20 CV of wash buffer I until baseline is reached. Change flow rate from 1 to 3 to 5 mL/min
when E
280
starts to drop.
Switch to wash buffer II (50 mM Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300 mM NaCl, 8 M,
20 mM imidazole, 2 mM β-mercaptoethanol, no need to readjust pH to 8.0.) for 6
CV .
Followed with wash buffer III (50 mM Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300 mM NaCl, 8
M urea, 50 mM imidazole, 2 mM β-mercaptoethanol) for 6 CV.
Then target peptide are eluted by Elute Buffer (50mM Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300
mM NaCl, 8 M urea, 300 mM imidazole, 2 mM β-mercaptoethanol, need to
readjust pH to 8.0). Wash column with water and, for long term storage, equilibrate in
20% ethanol and store at 4 °C. Dialyze eluted protein overnight against 5 L of 50 mM Tris
pH 8.0, using a Spectra/Por 3 Dialysis Membrane, MWCO 3500, 45 mm (cat. no.
S632724).

20
Cleavage:
Take out dialyzed solution and rinse membrane with 1-2 ml of dialysis buffer to get out all
protein. Measure MBP-TEV-CD33KQ concentration by UV-visible spectrometer (Agilent
8453). Blank against dialysis buffer. Add DTT to a final concentration of 1 mM and digest
with TEV (1:100) molar ratio, 1 TEV: 100 proteins. At 4 °C cleavage for 2h, collected
precipitate by centrifugation (10 min, 4600 rpm). Run a gel of supernatant. If there is still
uncleaved MBP , re-measure protein concentration of supernatant, add DTT and TEV again
to cleave remaining MBP fusion protein. If our target protein is hydrophilic peptide, there
won’t show any precipitate.
2
nd
Purification:
Charge 5 ml HiTrap IMAC HP column (cat no. 17-0409-03) with 5 ml of 0.1 M NiSO
4
at
2 ml/min. Then equilibrate HiTrap IMAC HP column with wash buffer I: 50 mM
Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300 mM NaCl, 25 mM SD S at 5 ml/min. Preparing the
sample with lysis buffer, let the final SDS concentration be 25mM. Load 3 ml of re-
suspended sample or supernatant in 5mL superloop and inject at 0.2 ml/min on column.
Collect peptide as flow through.
Be careful not to suck in air into solvent line. Elute bound proteins (fusion protein and
TEV) with 50 mM Sodium Phosphate pH 8.0, 300 ml NaCl, 300 mM imidazole. W ash
column with MilliQ-H
2
O and store in 20% ethanol.

HPLC:
Set Hamilton PRP-3 (305X7.0mm, cat. no. 79468) or Zorbax C18 column
temperature to room temperature or to 60 °C and turn on degasser.
Prime pumps with A, 0.1% TFA in MilliQ-H
2
O and B, 0.1% TFA, 100%
21
acetonitrile, using the program “prime_pumps_AA”.
If new solvent bottle has been used and air entered the solvent line, manual
purging has to be performed. Set injection valve to “inject” and equilibrate column
and sample loop with buffer A. Use programs “Load_AA”. Filter sample through a
Minisart SRP 15 0.20 μm filter. Set injection valve to “load” and load the sample.
Load no more than 5 ml into the 10 ml super-loop. To inject, set
injection valve to “inject”. If applicable, make multiple injections. Wait until
seeing the flow-through from each injection before making a new one.
Start desired gradient elution program. For example, gradient from 0 to 40% B in
20 min at 3 ml/min (program “0_40_20min_3ml_AA”). Adjust gradient start/stop
and slope to give a symmetrical and well separated peptide peak.
Collect peptide peak (can collect 1st and 2nd half separately). Turn off column
heater. Either run program “column_wash” (from 100% MilliQ-H
2
O to 50%
acetonitrile, 50% n-propanol in 30 min and back, followed by pump water storage)
or “store_pumps” for storage in water.
2.2.3 SDS phage analyzing
For TEV cleavage sample of MBP-CD33KQ, we use SDS phage to identify each fraction
of purification step.
Using ExpressPlus
TM
Page Gels (GenScript, Cat: M42012), the first lane loading
8µL Protein ladder standard, rest of the lane loading 20 µL sample + 5 µL SDS loading
dye (W/O DTT depending on need.) Using Tris-MOPS-SDS running buffer (GenScript,
Cat.No.M00138), run the gel at 140 volts for around 90 min.
SDS gel images were scanned by an Epson
TM
Perfection Photo V300 flatbed scanner.

22
2.2.4 NMR Sample Preparing
A: Freeze sample in liquid N
2
for 10-15 min and freeze-dry. On next morning, take
up freeze-dried powder in 67% CH
3
CN/33% H
2
O and measure peptide
concentration using the 500 μl cuvette. Vortex to help peptide dissolve.
B: Doing buffer exchange, using the centrifugal concentrator (GE Healthcare,
Vivaspin, 5kDa MWCO, 28-9323-59), exchange with buffer including 25mM
sodium phosphate pH7.4 and 120mM NaCl.(add TCEP to reduce Cys residues
if needed). Centrifugation under 4600 rpm, washing by 5mL exchange buffer
15min first, doing 1 to 10 exchange for at least 4 times.

23
3. Results

3.1 Purification of CD33(K283-Q364) by FPLC Chromatography
To purify His-tagged CD33(K283-Q364) from E. coli cell lysate, immobilized metal-
affinity chromatography (IMAC) was employed. Figure 6 depicts the chromatogram for
the purification of the fusion between maltose binding protein (MBP) and CD33(K283-
Q364). After loading the column (broad UV
280nm
absorbance), MBP- CD33(K283-Q364)
was eluted in 50 Na
2
HPO
4
/NaH
2
PO
4
pH 8.0, 300 mM NaCl, 8 M urea, 300 mM imidazole,
as can be seen by the sharp UV
280nm
peak reaching 1450 mV . Column flow-through and
eluted protein was analyzed by SDS-PAGE (Fig. 8).

Figure 6. FPLC of 1
st
His-tag Column purification of MBP-CD33KQ
The last sharp peak is MBP-CD33KQ, detection channel is 280nm. Resultant
signal = 1300 mv.
24

3.2 Cleavage of MBP-CD33(KQ) by TEV protease
To cleave MBP-CD33(KQ) by TEV protease immediately before CD33(KQ) to only leave
a singly Gly from the ENL YFQ-Gly TEV recognition site, we first used a molar TEV:
MBP-CD33(KQ) ratio of 1:50 in buffer 50mM Tris pH 8.0, 1 mM DTT.
Fig. 8. shows the results of the cleavage reaction carried out overnight. The molecular
weight of MBP-TEV-CD33KQ is 54 kDa. A band corresponding to this size
disappears in the course of cleavage to yield MBP alone (45 kDa). The molecular
weight of CD33 KQ is 9kDa but was not observed at the current cleavage
conditions. Thus, although at current conditions cleavage efficiency is essentially 100%,
CD33KQ may have been overdigested and degraded beyond detection. When performing
Figure 7. FPLC of 2
nd
His-tag Column purification of MBP-CD33KQ
Each peak is around 3mL sample injection, after cleavage the CD33KQ
extinction coefficient is pretty low (ε =2980 M
-
¹cm
-
¹).
25
a 2
nd
IMAC purification to remove His-tagged TEV and MBP , no proteins could be
detected in the the flow-through (Fig. 7) in support of an overdigestion of CD33KQ.

In order to slow the rate of TEV cleavage, we next tested cleaving MBP-CD33(KQ) at
room temperature and 4 °C using molar TEV-protein ratio of 1:100. Moreover, samples
were collected after 1h, 2h, 3h, 4h and overnight and frozen at -80°C until SDS-PAGE was
performed.
Figure 8.TEV Cleavage of MBP-CD33KQ From left to right, the lanes are protein
ladder, lysis of E.coli, first IMAC flow through, 1
st
IMAC elute before cleavage
(MBP-TEV-CD33KQ), after cleavage, 2
nd
IMAC binding fraction and 2
nd
IMAC
elute.

26

When loading the samples onto the gel, loading buffer without DTT was used. CD33KQ
contains one cysteine. When forming an intermolecular disulfide bond, a dimer of ~18kDa
is obtained, which is easier to detect. Fig. 9 shows that now CD33KQ could indeed be
detected and is best to cleave at 4°C for 2 hrs.

Figure 9. Time cause TEV cleavage MBP-CD33KQ results. The first lane
is before TEV cleavage. Group A is cleaved at room temperature, 1h, 2h, 3h,
4h and overnight. Group B is cleaved at 4°C, 1h, 2h, 3h, 4h and overnight.
27
3.3 NMR spectroscopy of CD33(Q233-Q364), GGH-CD33(V257-R291) and
COMMD3

In addition to MBP-CD33(KQ), I also investigated CD33(Q233-Q364). This construct
contains the transmembrane and cytosolic domains of CD33. It was purified analogously
to CD33(KQ) except that no overdigestion of the peptide took place. In the absence of
lipids, the transmembrane domain likely aggregated, which shielded the peptide from
overdigestions.
15
N-labeled peptide was reconstituted in 350 mM DHPC (write out
abbreviation), 105 mM DMPC (write out abbreviation), 25 mM Hepes pH 7.4 at a final
concentration of xx mM. Fig. 10 depicts the two-dimensional H-N correlation spectra of
CD33(Q233-Q364). The chemical shift dispersion conforms to a helical and unfolded
structure for the transmembrane and cytosolic domains, respectively.
To assess the aggregation state of the transmembrane domain yet another construct was
prepared analogously. The transmembrane domain alone, residues xxx-yyy, was preceded
by Gly-Gly-His, which is referred to as the A TCUN motif (ref). ATCUN binds divalent
cations and can be used to incorporate paramagnetic Mn
2+
to assess aggregation with wild-
type xxx-yyy by NMR spectroscopy. To test the feasibility of this approach, we recorded a
two-dimensional H-N correlation spectra of
15
N-labeled GGG-CD33(xxx-yyy) in 350 mM
DHPC, 105 mM DMPC, 25 mM Hepes pH 7.4. Fig. 11 shows the spectrum with a peak
distribution typical of a helical transmembrane domain and uniform signal intensities
characteristic of a well-folded protein.
Finally, the cytosolic CD33 domain is predicted to interact with the protein COMMD3. To
be able to test this hypothesis, Alan Situ in our laboratory purified COMMD3 and we
recorded a two-dimensional H-N correlation of the
15
N-labeled protein to assess if it is
folded. Figure 12 shows a spectrum characterized by a chemical shift dispersion and signal
intensities that clearly indicate the folded nature of COMMD3. Thus, interactions studies
between COMMD3 and CD33KQ will be feasible.
28

Figure 10.NMR Spectrum of CD33 TM domain and cytosolic domain.
The TROSY-type H-N correlation spectrum of 0.5 mM
15
N-labeled CD33QQ peptide in 350 mM
DHPC, 105 mM DMPC, 25 mM Hepes pH 7.4 was recorded at 35 °C and 700 MHz.

29

Figure 11.NMR Spectrum of GGH-CD33V257-R291. The TROSY-type H-N
correlation spectrum of 0.5 mM
15
N-labeled CD33 peptide in 350 mM DHPC, 105 mM DMPC, 25
mM Hepes pH 7.4 was recorded at 35 °C and 700 MHz.

30

Figure 12. NMR spectrum of COMMD3. The TROSY-type H-N correlation spectrum of 0.5
mM
15
N-labeled COMMD3 in 25 mM Hepes pH 7.4 was recorded at 35 °C and 700 MHz.

31
4. Discussion & Conclusion

4.1 Choice of CD33 fusion protein
We have tested two different fusions proteins for expressing CD33 (chart 1), GB3 and
MBP . With the highly soluble GB3 high expression levels of my target peptide were
obtained. Specifically, for GB3R2-TEV-CD33YQ, I obtained pure CD33YQ after TEV
cleavage (Figure 10). However, when I used this tag to produce the cytosolic domain of
CD33, either as CD33GQ or CD33KQ, it was difficult to separate GB3 from CD33
peptides after TEV cleavage. This was partly due to the similar molecular weights of
GB3R2 and CD33KQ of approximately 9kDa. Because of this similarity, it was difficult to
identify whether CD33 peptides exist in SDS P AGE. Moreover, both GB3R2-TEV-CD33
and GB3 seemed to co-precipitate with CD33 following TEV cleavage and I was unable to
separate these species in subsequent purification steps. I therefore tested MBP as the fusion
protein of CD33KQ. Its larger size and higher solubility eliminated similar problems,
allowing the successful purification of CD33 K283-Q364 (Figure 9).
The CD33 transmembrane domain (CD33V257-R291) preceded by the GGH motif was
also purified as fusion to MBP . Its high quality NMR spectrum will allow the evaluation of
the aggregation state of the CD33 transmembrane domain.
4.2 Degradation of the CD33 cytosolic domain by TEV protease
Even when using the MBP-TEV-CD33KQ construct, I was initially puzzled by the
apparent disappearance of the approximately 9kDa SDS-PAGE band corresponding to
CD33KQ. Moreover, no such band could be detected after removing TEV protease and
MBP by IMAC. I therefore hypothesized that CD33KQ was degraded by TEV protease
32
(overdigestion). Figure 13A depicts the canonical cleavage sequence of TEV . Inspection of
the CD33KQ sequence highlights a sequence that exhibits some similarity to this sequence
(Figure 13B). Specifically, EXLXY and FHG may get cleaved. X denotes any amino
acids.

In support of this hypothesis, lowering the cleavage temperature and shortening cleavage
times, yielded the desired approximately 9 kDa CD33KQ product. Under these conditions,
cleavage efficiency is approximately 40%, i.e., a trade between yield and avoiding
degradation is necessary. While my purification strategy is successful, an alternative may
be offered by SUMO and SUMO protease Ulp1 that recognized the 3D structure of
SUMO and cleaves in a sequence-independent manner.

Figure 13. Comparison of TEV cleavage site&CD33KQ Sequence. A
is the model of TEV cleavage processing, the binding and cut amino
acids sequencing is ENLYQ//(S/G). B is the sequence of CD33KQ, the
yellow highlight part is the probable mimic cleave site might be attacked
by TEV.
33
4.3 Future directions
At this stage, we have pure CD33KQ, the cytosolic CD33 domain, and the putative
COMMD3 ligand. This allows a quantitative test of the interaction of the cytosolic CD33
domain and COMMD3 by isothermal titration calorimetry (ITC).
Subsequently, we aim to purify SHP-1, the first discovered CD33 ligand, which controls
downstream signaling and also determine its binding affinity to CD33. Comparing
affinities between COMMD3 and SHP-1 will provide further insight into the assembly of
CD33 signaling. Finally, additional ligands can be studies to fully understand the
modulation of CD33 signaling pathways.
The availability of the GGH-tagged CD33 transmembrane domain now also allows the
immediate assessment of its aggregation state. If mixing paramagnetic with diamagnetic
CD33 produces NMR line broadening of the diamagnetic species, paramagnetic relaxation
induced-broadening of the
1
H-line may occur as a result of aggregation.

34
5. References

[1] Jiang, T., Y u, JT., Hu, N. et al. Mol Neurobiol (2014) 49: 529.
https://doi.org/10.1007/s12035-013-8536-1

[2] Chapter 7, World Alzheimer Report 2018, Alzheimer’s Disease International. 2018

[3] Ana Griciuc, et al. Alzheimer’s Disease Risk Gene CD33 Inhibits Microglial Uptake of
Amyloid Beta, Neuron. 2013 May 22; 78(4): 631–643.
[4] Frank S. Prato, William F . Pavlosky. et al. Screening for Dementia Caused by
Modifiable Lifestyle Choices Using Hybrid PET/MRI, Journal of Alzheimer’s Disease
Reports 3 (2019) 31–45
[5] Types of Dementia, Alzheimer's Association’s official website, 2019

[6] Chitnis T, Weiner HL. CNS inflammation and neurodegeneration. J Clinical
Invest. 2017 Oct 2;127(10):3577-3587.
[7] Ricardo B. Maccioni, Andrea González, et al. Alzheimer´s Disease in the Perspective of
Neuroimmunology, Open Neurology Journal. 2018; 12: 50–56.
[8] Leslie Crews, Eliezer Masliah. Molecular mechanisms of neurodegeneration in
Alzheimer's disease, Human Molecular Genetics, V olume 19, Issue R1, 15 April 2010,
Pages R12–R20.
[9] Tanzi RE (2013) A Brief History of Alzheimer's Disease Gene Discovery . J Alzheimers
Dis 33: S5-S13.
[10] Mario F . Mendez. Early-Onset Alzheimer’s Disease, Neurologic Clinics. 2017 May;
35(2): 263–281.

[11] Guerreiro RJ, Gustafson DR, Hardy J. The genetic architecture of Alzheimer’s
disease: beyond APP PSENs and APOE. Neurobiology Aging 2012; 33:437–56.
[12] Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW , et
al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late
onset families. Science (New Y ork, NY) 1993;261:921–3.
35
[13] Ginhoux F ., Lim S., Hoeffel G., Low D., and Huber T. 2013. Origin and
differentiation of microglia. Front. Cell. Neurosci. 7:45 10.3389/fncel.2013.00045

[14] Sujatha P . Paul, Lynn S. Taylor , Eryn K. Stansbury and Daniel W . McVicar. Myeloid
specific human CD33 is an inhibitory receptor with differential ITIM function in recruiting
the phosphatases SHP-1 and SHP-2. Blood 2000 96:483-490;

[15] Crocker PR, Paulson JC, V arki A. Siglecs and their roles in the immune system. Nat
Rev Immunol. 2007 Apr; 7(4):255-66.

[16] Ravetch JV , Lanier LL. Immune inhibitory receptors. Science. 2000 Oct 6;
290(5489):84-9.

[17] Biswajit Khatua, Saptarshi Roy, and Chitra Mandal. Sialic acids siglec interaction: A
unique strategy to circumvent innate immune response by pathogens. Indian J Med Res.
2013 Nov; 138(5): 648–662.

[18] Arora, A. (2012). Solution NMR Spectroscopy for the Determination of Structures of
Membrane Proteins in a Lipid Environment. Lipid-Protein Interactions, 389–413.

[19] Roland B. W alter, Frederick R. Appelbaum, Elihu H. Estey, Irwin D. Bernstein. Acute
myeloid leukemia stem cells and CD33-targeted immunotherapy. Blood. 2012 Jun 28;
119(26): 6198–6208.

Abstract (if available)

Abstract The molecular origins of Alzheimer’s Disease (AD) onset and progression are still incompletely revealed. A major accepted cause of AD is the aggregation of fibrotic amyloid peptides (Aβ) in the human brain. We do not have any efficient therapy to treat AD for now. Genetic studies show that Cluster of differentiation 33 (CD33) is a risk factor for AD [1]. This transmembrane receptor binds a sialic acid binding site and resembles an immunoglobulin-like lectin. CD33 can regulate innate immunity, but its functions in the brain still remain unknown. ❧ Alzheimer’s Disease is one of the most prevalent chronic neurodegenerative disease, the most common cause of dementia and the sixth leading cause of death in the United States. From the world report of Alzheimer’s disease 2018, there is 50 million people worldwide living with dementia. By using some statistic model, the evaluation of people who living with dementia will more than 152 million by 2050 [2]. CD33 is located in the plasma membrane of microglia membrane and can influence microglia phagocytosis of Aβ, which alleviates Alzheimer’s Disease [3]. ❧ Multidimensional heteronuclear NMR spectroscopy is a major technique for determining the structure of bio macromolecules in solution. In my thesis, I started to examine the CD33 structure by multidimensional, heteronuclear NMR spectroscopy. Specifically, the transmembrane domain and intercellular domain, which includes a SHP (Src Homology Phosphatase) binding site, were examined. Also we will use Paramagnetic-Tag to label TM domain to study the aggregation state of the TM domains, which may contribute to cell signaling. After finishing the structure work, we will move to use ITC (Isothermal Titrate Calorimetry) to characterize the binding affinity between the cytosolic domain of CD33 and some candidate protein. Our work aims at finding some molecules which have the potential to accelerate the microglia phagocytosis to treat AD.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Expression and purification of extracellular domain of human CD33 in Escherichia coli and Pichia pastoris

PDF

Molecular basis of CD33 receptor function, and effects of a destabilizing transmembrane motif on αXβ2 integrin

PDF

Interaction of monoclonal antibodies MW1 and PHP1 with huntingtin exon1 protein

PDF

Structural studies on functional amyloids and the mechanism of aggregation and disaggregation of Huntingtin Exon 1

PDF

Role of mitochondrially derived peptides in the inhibition of IAPP misfolding

PDF

Dual effects of transmembrane proline residues on integrin function

PDF

Facilitating unambiguous NMR assignment by solid–state NMR using segmental isotope labeling through split-inteins

PDF

Understanding the role of APP and DYRK1A in human brain pericytes

PDF

The structure and function of membrane curving proteins on different membrane shapes and their regulation by post-translational modifications

PDF

Uncovering the influence of N-terminal phosphorylation on conformational dynamics of huntingtin exon 1 monomer

Asset Metadata

Creator Wu, Mingke (author)

Core Title Role of CD33 structure and function in Alzheimer’s disease

School Keck School of Medicine

Degree Master of Science

Degree Program Biochemistry and Molecular Medicine

Publication Date 07/30/2019

Defense Date 05/09/2019

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

Tag aggregation state,Alzheimer's disease,binding affinity,CD33,multidimensional heteronuclear NMR spectroscopy,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ulmer, Tobias (committee chair), Langen, Ralf (committee member), Siemer, Ansgar (committee member)

Creator Email mingkewu@usc.edu,mingkewu_usc@163.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-201604

Unique identifier UC11663295

Identifier etd-WuMingke-7698.pdf (filename),usctheses-c89-201604 (legacy record id)

Legacy Identifier etd-WuMingke-7698-0.pdf

Dmrecord 201604

Document Type Thesis

Format application/pdf (imt)

Rights Wu, Mingke

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