Biomaterials, the Nanobiointerface, and Surface Modification Strategies Buddy D. Ratner University of Washington Engineered Biomaterials (UWEB), Department of Bioengineering, Department of Chemical Engineering, University of Washington,

[email protected] An NSF Engineering Research Center

Nanotechnology (and nanoscience) continue to grow in prominence

Where did nano/bio ideas come from? How are they relevant to biocompatibility? How can we use nano ideas for biomaterials? Biointerface and biosurface

The roots roots of of nanotechnology: nanotechnology: The 1770

Ben Franklin

1890

Agnes Pockels

1920

Irving Langmuir / Katherine Blodgett

1940+

Pauling / Watson & Crick

1959

Richard Feynman

1970

Helmut Ringsdorf

1986

Binnig and Rohrer

2000

Bill Clinton

Modern nano ideas

The principles of physics, as far as I can see, do not speak against the possibility of maneuvering things atom by atom. It is not an attempt to violate any laws… Richard Feynman, 1959 What went on in the intervening years?

Imagine the possibilities… Bill Clinton, CalTech, 2000

The principles of physics, as far as I can see, do not speak against the possibility of maneuvering things atom by atom. It is not an attempt to violate any laws… Richard Feynman, 1959 The molecular biologists and molecular bioengineers have been busy! (1960s - 2000)

Imagine the possibilities… Bill Clinton, CalTech, 2000

Rube Goldberg Simplified Pencil Sharpener

Open window (A) and fly kite (B). String (C) lifts small door (D) allowing moths (E) to escape and eat red flannel shirt (F). As weight of shirt becomes less, shoe (G) steps on switch (H) which heats electric iron (I) and burns hole in pants (J). Smoke (K) enters hole in tree (L), smoking out opossum (M) which jumps into basket (N), pulling rope (O) and lifting cage (P), allowing woodpecker (Q) to chew wood from pencil (R), exposing lead. Emergency knife (S) is always handy in case opossum or the woodpecker gets sick and can't work.

Biological Triggers of Cell Functions hormone

(more than 1000 such G protein receptors have been identified) Adenylyl Cyclase

G protein receptor

cell membrane

hormone binding triggers conformational change

cytosol

G protein

activated G protein ATP cAMP an important second messenger

Nature: the ultimate nanoengineer

Soong, et al. Science, 290, 1555 (2000)

Schematic diagram of the F1-ATPase biomolecular motor-powered nanomechanical device. The device consisted of (A) a Ni post (height 200 nm, diameter 80 nm), (B) the F1-ATPase biomolecular motor, and (C) a nanopropeller (length 750 to 1400 nm, diameter 150 nm). The device (D) was assembled using sequential additions of individual components and differential attachment chemistries.

Ned Seeman - DNA architechures

Artist’s conception

Marlaria protozoa invade red cells in a similar manner

“Rules” common to nanotechnology and molecular biology • “Bottom up” fabrication (molecules up) • self assembly • supramolecular structure • self replication (J. Rebek, etc) • a new applied physics - quantum tunneling - diffusions and flows

• integrated use of soft and hard matter • engineering with kT noise • surface science principles • hydrophobicity / interfacial properties

Scaling relationships - nano to macro µm flow Protein

DNA

20µm 10-9 m

10-9 m - 10-5 m

20µm Uncoated

Coated 10-5

10 m

Implant device Array device

Nanogen

10-5 m - 10-2 m

10-2 m

So, how does “nano” connect to “macro-biomaterials?”

Coat the device with a nanometerdimension film.

Coat the device with proteins.

USING NANO CONCEPTS IN DESIGNING BIOMATERIALS Introduction to modern biomaterials -How can nanotechnology help? Surfaces of biomaterials

In 50 years since the first biomaterials (as we know them today) were developed, the field has evolved into a $100 billion endeavor that saves lives and improves the quality of life for millions.

B-17

Common Medical Implants Finger joint

Breast implant

Heart valve

CNN

Hip joint

Artificial heart

http://www.usc.edu/dept/biomed/ bme490.981/artificial_heart.htm

Intraocular lens (IOL)

Success in Biomaterials after ~50 years of research and development: W O R L D W I D E

• IOLs (>7,000,000/yr) (PMMA, silicone) • Hip and knee Prostheses (>600,000/yr) (titanium, steel, PE) • Vascular Grafts (>300,000/yr) (Teflon, Dacron) • Heart Valves (>200,000/yr) (carbon, fixed tissue) • Percutaneous Devices (>100,000/yr) (titanium, silicone) • Stimulatory Electrodes (>100,000/yr)(platinum, iridium) • Catheters (millions/yr)(silicone, PVC, PEU, Teflon) • Stents (>2,500,000/yr)(stainless steel) U.S. healthcare market (1998) > $1 trillion

Millions of lives saved / The quality of life improved for millions more

how well do biomaterials really work? • IOLs (25%-50% reoperation rate) • Hip and knee Prostheses (still a 10-15 yr lifetime) • Vascular Grafts (no healing) • Heart Valves (calcification or thrombosis) • Percutaneous Devices (no seal) • Stimulatory Electrodes (electrode encapsulation) • Catheters (thrombosis, infection; 1,000’s of deaths/yr) • Stents (clotting and closure) • Contact lenses (discomfort and eye injury) • Dental Implants (loosening)

Problems of much concern, but also, An opportunity…

Why the complication rate? What can we do about it?

A central premise of the biomaterials field has been:

The surface dictates the biological reaction

BIOMATERIAL

How does the surface dictate healing?

Biocompatible Biomaterials Healing: Encapsulation and Isolation 1.

8.

Biomaterial

injury

The biomaterial is encapulated and isolated

Biomaterial Biomaterial

2. protein adsorption neutrophil

fibronectin laminin GAG TSP? SPARC?

ECM: collagen

Biomaterial

7.

3. cell interrogation

fibroblast

monocyte / macrophage

4. release

In conjunction with ECM molecules...

6.

Communication to fibroblasts

release

Growth and attachment inhibitors and stimulators

PDGF FGF interleukins

TGF or EGF TNF

5.

Giant cell formation at the Implant surface -- frustrated phagocytosis

SEM of tissue surrounding subcutaneous device in rat: 6 weeks after implant PTFE device loose connective tissue-

From K. Ward, Isense Corp

dense foreign body capsule

e th n k eo s a rg su

An operational definition of “biocompatibility“ The foreign body reaction

no adhesion between implant and the capsule

circa, 2001 thin walled capsule

Ti

PET

PEU

Implant

C PLA

PDMS PE f f o ed body l l Wa the m fro The implant, after approximately 1 month, is found within a thin, relatively acellular, collagenous sac. The reaction site is quiescent.

What is the similarity between all these widely different materials? Ti

PDMS

PEU

PET

C

PLA PE

They all have uncontrolled interfacial proteins! • HYPOTHESIS •

Can we conclude that surface properties do not matter? Not a good conclusion: • Biology uses surfaces and interfaces, too. • All in vitro biomaterials directly exploit surfaces

Nature does it’s work at interfaces and surfaces: e c a

Lipid membrane

f ur

s

e ul

(Proteins Sugars Water)

cell

ec l o

m

Collagen and other extracellular matrix proteins

Medical device (biomaterial)

(the molecules assemble to form a surface) www.cellsalive.com

http://cellbio.utmb.edu/cellbio/membrane_intro.htm#Architecture

Biology’s Surface Tricks 1. Complexity 2. Recognition • • • •

enzyme-substrate antibody-antigen DNA-RNA-protein lectin-carbohydrate

3. Assembly (order) • collagen • cell wall • supramolecular structure

4. Mobility 5. Optical sense

Surface Modification: the rationale surface

Unchanged:

bulk

- mechanical properties - configuration - surgeon handling characteristics - manufacture Only alter the surface zone to influence: biocompatibility other performance parameters

Surface Modification Methods • Plasma treatment and deposition • radiation grafting • chemical reaction of the surface • ozonolysis • photoreaction • ion implantation • ion etching • solvent cast films • surface active modifiers (low and high MW) • metalization • self assembly • micro-contact printing • immobilization of biomolecules

UWEB How biomaterials heal now!

Biomaterial

The foreign body reaction

Biomaterials healing in the future

Biomaterial

A reconstruction of the anatomy!

Normal wounds heal this way. Why not our biomaterials?

University of Washington Engineered Biomaterials (UWEB)

UWEB Vision Statement ... ... to to evolve evolve engineered engineered biomaterials biomaterials that, that, by by emulating emulating nature’s nature’s own own mechanisms, mechanisms, control control with with precision precision the the interaction interaction of of biology biology with with synthetic synthetic materials. materials. This This leads leads to: to: •• biomaterials biomaterials that that heal heal and/or and/or function function in in an an improved improved manner manner •• improved improved diagnostics diagnostics •• aa new new intellectual intellectual frontier frontier as as aa platform platform for for our our educational educational and and outreach outreach efforts. efforts. Can mechanistic biology be reduced to a clockwork model (the province of the engineer)?

an NSF Engineering Research Center

University of Washington Engineered Biomaterials (UWEB)

research

team effort

industry

education

UWEB is…

• Biomaterials that heal (engineered biomaterials) • Improved diagnostics 20 professors and 100 students focused on biology at surfaces twenty six companies in partnership to advance biomaterials a focused effort to bring more students, and more diverse students, into engineering and science

So how can we get around this foreign body reaction?

The Central UWEB Strategy •stealth materials and non-fouling materials Use where necessary

•Prevent non-specific interactions Always!

•Encourage specific interaction Engineered surfaces

An Engineered Biomaterial • the biology is well understood (the correct receptor interaction) • molecules in defined orientation, conformation stabilized • bland, non-interactive regions between receptors? (pattern the surface chemistry)

10 nm Medical device

A focus on the interface:

The Basic Repertoire of Surface Analysis Methods x-rays

. . . . . ... electrons

primary ions + + +

ESCA

SIMS .. .. ..-secondary .. ions +

Macintosh II Macintosh II

laser

AFM

detector

θ lever

Macintosh II

Contact angle

H

T AL

E ST A strategy for non-fouling (stealth) surfaces

Poly(ethylene glycol) or Poly(ethylene oxide) (PEO) (O-CH2CH2)n

PEO surfaces are found to resist protein and cell pickup Our monomers are glymes, e.g. tetraglyme: CH3 (O-CH2CH2)4 O-CH3

A 20 nm coating

Advantages of Plasma Deposits • • • • • • • • • • •

dry processing rapid pin-hole free conformal can be done on continuous basis tenacious adhesion to the substrate many possible chemistries many possible substrates monomer costs are negligible sterile product much precendent

Plasmas -- Unique Monomers and Unique Surface Chemistries Thin film deposition “Monomer” Methane Tetrafluorethylene Benzene Methanol Ethylene oxide Tetraglyme Acrylic acid Allylamine Hydroxyethyl methacrylate N-vinyl pyrrolidone mercaptoethanol

Film Characteristics Hydrocarbon, diamond Fluoro, hydrophobic graphitic polar Rigid, polar Non-fouling -COOH-rich Amine-rich Hydrogel, hydroxyl hydrogel Sulfur-containing

H

Fibrinogen Adsorption to Plasma Polymerized Tetraglyme on FEP

T AL

E

ST

Absorbed Fbgn (ng/cm

2

)

160 140

UWEB 116

120 100 80 60

Teflon control

Various glyme coatings

40 20 0

0 A

1.6

3.7

6.3

6.4

B

C

D

E A (EtOH)FEP

0

Data of T. Horbett, V. Pan and B. Ratner

Pseudomonas aeruginosa attachment and growth

Growth Mode: Net Accumulation of colony forming units (CFU) 160 140

Glass

Glymes

Cyclics

120 % change in number of CFU

100 80 60 40 20 0 Gls 1 Gls 2

Mono 1 Tri 1 Tri 2 Tetra Tetra Diox 12cr4 15cr5 15cr5 5w 20w 5w 20w Data of E. Johnston, J. Bryers and B. Ratner

Photopatterning of RFGD Films UV

Cast

Expose

Develop

UV

Re-expose

Analyze

Re-develop

Polymerize

Methylobacterium extorquens AM on patterned tetraglyme

Upon seeding

70 hrs. later

Data of Yael Hanein, Karl Boehringer, Vickie Pan and Buddy Ratner

Methylobacterium extorquens AM on patterned tetraglyme 0

70 hrs.

Images taken every 2 hrs. Data of Yael Hanein, Karl Boehringer, Vickie Pan and Buddy Ratner

2

Monocyte Density (cells/mm )

Monocyte adhesion to plasma polymerized tetraglyme 2000 R2=0.983 1500

1000

500 5, 10, 20 W

0

0

50

100

40, 60, 80 W; FEP

150

200

250

300

350

400

Adsorbed Fibrinogen (ng/cm2) Mingchao Shen, et al.

Histology of tissues surrounding implants Tetraglyme Implant FEP Implant

60 50 Fibrous 40 Capsule 30 Thickness (um)20 10 0

FEP Tetraglyme

Skin side

Muscle side

Collagenous Tissues

M. Shen, et al.

Conclusion

We believe non-fouling surfaces to be necessary, but not sufficient, for biomaterials that heal.

UWEB

The Clues to Healing

Materials that “heal” without a capsule... Hydroxyapatite (bone) 5 µm porous structures fine fibers Kukobo Titanium (bone) Tyrosine polycarbonate? RGD peptides?

Molecules always present in healing wound sites SPARC Osteopontin Thrombospondin Fibronectin Fibrinogen Laminin V

UWEB “Rosetta Stones” -- read the code and learn how nature heals!

5 µm pores Made by microsphere templating (A. Marshall)

Porous materials have been examined for toxicology and endotoxin

Crystalline surface of packed 60µm beads (oblique view)

Porous pHEMA templated with a crystalline array of 60µm beads

Electrospinning of fibro-porous materials “skinny fibers”

Capsule Thickness µm (mm)

55 45 35 25 15 5 -5 0

5

10

15

20

25

FiberDiameter(mm)µm

Joan Sanders, et al.

Turn on specific reactions

Matricellular Proteins

ECM

SPARC Tenacin Osteopontin Thrombospondin

Cells

Osteopontin coating is a normal part of implant healing: ePTFE subcutaneous implants 7 days

28 days

Osteopontin stains brownish-red Giachelli, et al.

Osteopontin-Immobilized PolyHEMA (Lysine-Immobilized control)

50 m

50 m Lysine-Immobilized PolyHEMA, explanted after 28 days

OPN-Immobilized PolyHEMA, explanted after 28 days

BV Score = 1

BV Score = 3 Data of Stephanie Martin, Cecelia Giachelli and Buddy Ratner

Vascular density and capsule thickness of foreign body capsules in control and TSP2-null mice (4 week implantation)

genotype Thrombospondin normal mice

Thbs2+/+

Thrombospondin knockout mice

material

vessels/capsule

Capsule thickness facing dermis (µm)

body wall (µm)

PDMS

10±10

55±8

24±3

Ox-PDMS

12±8

58±7

25±4

PDMS

100±20

91±7

36±5

Ox-PDMS

90±14

95±8

38±5

Thbs2-/-

Data of Kyriakides, Leach, Hoffman, Ratner and Bornstein Proc. Natl. Acad. Sci. USA 96, 4449-4454.

How can we deliver these “healing” signals on real world medical devices? Molecular templating is a possibility...

Proteins on mica CH 2O H O

Coat with sugar Deposit plasma film

Glue to glass

O H H O

O O H

HOC H2 O H O O H

O H

CF3 CF CF2 CF CF CF2 CF CF CF CF2 C CF3 CF2 CF CF

Peel off mica/ dissolve proteins

Do the pits recognize the template protein?

Galen Shi, et al. Nature, 398, 593-597(1999)

A hypothesis for the recognition mechanism

Imprinted Binding Pocket CHO O

H

HOC O CH 2

O

O OH HO

OH O OH

HO

HO OH

O OH

H

H

OH

H

N

HOCH O

OH

Ser

O

OH HO

O

OH

O

Lys

Glu

Template Protein

O HO

O OH

N

CH

OH

H

N

His

Tyr

HO

OH O

H

O H O

CH

HO

O H

N

Asn

Examples illustrating the fidelity of the imprint to reproduce the template template

TEM image of the imprint

AFM

3 nm

1 m

E. coli

Glutamine Synthetase (MW 600 KD)

14 nm

20 nm

E. coli

1 m

Distribution (%)

imprint 20 15 10 5 0 5

10 15 Diameter (nm)

20

Surface Topography (AFM) of Protein Imprints 45 nm

4 nm

6.5 nm

14 nm

5 nm

3 nm

Albumin

Fibrinogen

(MW 66 KD)

(MW 340 KD)

0 20 nm

0 20 nm 200

200 0

0 400

200 400

BSA IMP

600 nm

400

200

Fbgn IMP

400 600 nm

Microcontact printing (µCP) silicon

coat with resist and expose

Silicon

resist

Silicon

Silicon PDMS

Silicon

PDMS

develop resist

Silicon PDMS

etch

Silicon PDMS

strip

Silicon

silanize coat with PDMS strip from silicon ink the stamp protein thiol silane polymer

stamp a surface

Visualization of Protein Recognition: AFM BSA

SA monolayer Stamp

PDMS

PDMS

Stamp

Mica

SA stamping SA Biotin-10nm Au

AFM

Mica

BSA backfill Adsorbed protein Pattern

?

Biotin-BSA-Au Incubation

Template imprinting

Imprint

Imprint

Imprint

SA

SA/BSA co-adsorption

H Shi, B. Ratner, M. Garrison, Pat Stayton, Sandro Ferari

10 nm Au-biotin bound to adsorbed SA - AFM 150

100 nm 20 0 10

0 50

20

10 20 30

0

50

100

0 150

m

30 40

m

40

Nature, 398, 593-597(1999)

The proteins on mica are disordered -- this is a sub-optimal template! binding site Now

Desired

Imprint

Question: Control at the nanoscale?

binding site

How will we analyze ordered proteins at interfaces?

• static secondary ion mass spectrometry

SIMS primary ions + + +

.. .. ..-secondary .. ions +

Macintosh II

+ Static SIMS spectrum of a fibrinogen film counts/10 2

3 P, R 2.5 V, D

2 1.5

S

F, W

1 0.5 0 50

E, K H

L

H D M

70

Y

F

90

R

R M E

m/z

110

F

W Y

130

150

Each letter indicates an amino acid Mantus and Ratner

Y

PC

PCA Principal Components Analysis •PC1: direction of the greatest variance •PC2: orthogonal axis defining the next greatest of variance •Scores: projection of the samples onto the new PC axes •Loadings: direction cosines of the matrix rotation

2

......... . ....................... ... ..

1 C P

matrix rotation

X

PCA Distinguishes ToF-SIMS Spectra 0.06 PC 2 (19%)

Hemoglobin IgG

0.04 Fibronectin

γ-Globulins

0.02

Myoglobin

BSA

0

Papain Transferrin

-0.02

Collagen

Lactoferrin Fibrinogen

-0.04

Cytochrome c

Lysozyme

-0.06 -0.08 -0.15

-0.1

-0.05

0

0.05

0.1

PC 1 (53%)

Proteins adsorbed onto mica from 100 µg/mL solutions Matt Wagner and Dave Castner

Controlling and measuring antibody orientation Effect of pH on protein charge distribution Isoelectric point of IgG1, their fragments, and hCG+ IEP of Fc Isoelectric Point (IEP) of antibody

IEP of Fab

Mouse monoclonal anti-hCG

I.E.P. 6.8

Fab Fc

8.5 6.1

+ 2

+ 2

dipole

Dipole moment

Fab

Fab

- 2 2

+

+

+

+

Human chorionic gonadotropin (hCG)

Fc Hua Wang, S. Jiang

Linking SIMS spectra to protein structure • Information on anti-hCG structure from the protein data bank can tell us relative composition of amino acids in Fab and Fc:

Amino acid composition ratio Fab/Fc 3.5

TYR 3

2.5

GLY SER

2

ARG LEU 1.5

THR 1

0

CYS

VAL

ASP ILE 0.5

ALA MET

GLN

PRO

GLU LYS

T R P

ASN PHE HIS

The SIMS Experiment fab fragment

anti-hCG on C11NH2 N N N N N N N N N N H2 H2 H2 H2 H2 H2 H2 H2 H2 H2

anti-hCG on Au (111)

fc fragment

anti-hCG on C15COOH H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

Multivariate analysis of SIMS data scores plot from PCA of Fab and Fc fragments

PC2(16.32%)

0.03 0.02 0.01 0 -0.01 -0.02 -0.03 -0.04

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

0.04

P C 1 ( 7 9 . 5 7 % )

fc fragment on Au

fab fragment on Au

Loadings

Loadings plot for PC1 (67.94%) form PCA 0.8 0.6 0.4 0.2 0 -0.2 -0.4

86.097: Leu/Ile 70.0668: Val, Pro, 84.084: Leu/Ile, Lys Arg 72.08: Val 44.05: Ala, Lys 60.45: Ser 44.013: Asn 59.049: Arg

120.08: Phe 110.07: His

73.06: Asn 74.06: Thr 72.047: Ala, 71.01: Ser Thr

87.06: Asn

147.04: Tyr

m/z

Hua Wang, S. Jiang

Applying the PCA model to anti-hCG

anti-hCG on C11NH2

sco res plot from PCA of ant i-hCG 00 .04

PC2(16.32%)

00 .03 00 .02 00 .01

anti-hCG on Au (111)

0 -00 .01 -00 .02 -00 .03 -00 .04 -00 .05

00 .1

00 .15

00 .2

00 .25

00 .3

00 .35

00 .4

PC1 (79.57%)

ant i-hC G on C 11NH2 S A M s ant i-hC G on Au sur face

anti-hCG on C15COOH

ant i-hC G on C 15C OOH SAMs

PCA analysis of the SIMS spectra show differences among anti-hCGs with different orientations.

dipole

Fab

Fab

N N N N N N N N N N H2 H2 H2 H2 H2 H2 H2 H2 H2 H2

Fc

IgG molecule H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

H O O C

Using this alignment, monoclonal antibodies to specific domains on osteopontin can be used to align the OPN. These can then be used as templates for imprinting.

data from Hua Wang (in collaboration with Shaoyi Jiang, Chemical Engineering Department)

Adenovirus Nature knows how to order proteins and deliver protein signals.

Adenoviruses are icosahedral particles. The capsid (protein coat) is built up from 252 capsomers (T=25), of which 240 are hexavalent and 12 (situated at the apices) are pentavalent. A "penton fibre"projects from each apex. http://www.uct.ac.za/depts/mmi/stannard/adeno.html

What are the implications of template recognition surfaces? Immobilized biomolecules on “real” medical devices have the following limitations:

Hip prosthesis

Coat with nanopits for a key recognition protein

Implant in bone

The device concentrates the body’s protein

•Low stability •Expensive •Hard to sterilize •Prone to contamination •Complex regulatory position

Heals into bone

A sugar-coated medical device. The patient’s own proteins are used for healing.

The Hallmarks of an

Engineered Biomaterial 1. We know which reaction we want to control 2. We know the recognition events to control it 3. We will inhibit non-specific interactions

Nano and Surface skills that will be needed: non-fouling, immobilization, orientation, mimic, pattern, analysis

WHAT WILL THE FUTURE HOLD?

Biomaterials (2001-2020) Precison nano-surfaces (2010-2040 Tissue Engineering (2005-2050) Regenerative medicine (2020+ ) “The art of medicine consists in amusing the patient while nature cures the disease.” Voltaire (1694-1778)

J. Donoghue, Brown University

UWEB 2002