1. 1 e. Piezzoelectric methods Based on the change in the frequency of vibration of a piezoelectric crystal when the target complementary sequence hybridizes with the immobilized probe. The quartz crystal microbalance (QCM) is the most commonly used device, which comprises a piezoelectric quartz crystal with a determined frequency of oscillation under an applied potential. When the oligonucleotide is immobilized on the surface, the frequency of oscillation changes proportionally to the immobilized mass. Advantages: sensitivity and time-resolution. Sub-nanogram can be measured and it allows real-time measurements, thus facilitating the determination of association, binding and dissociation rate constants Multi-array format: uneasy. Analyst, 2014,139, 1576-1588 Basic device More modern devices : cantilevers

2. 2 f. Electrochemical methods Electrochemical transduction is simple, of low instrumentation costs and high sensitivity (comparable to fluorescence). Several electrochemical methods : Electroactive hybridization indicators (most common) Capacitance Enzymatic labels … Electroactive hybridization indicators These indicators are cationic metal complexes, such as [Co(bpy) 3 ] 3+, [Co(phen) 3 ] 3+ and [Ru(bpy) 3 ] 3+, or organic compounds, such as Hoechst 33258 and daunomycin, that recognize the DNA helix and intercalate selectively into double-stranded DNA (dsDNA). Nature Biotech. 2003, 21, 1192 - 1199 PNAS 2003, 100, 9134–9137 Oxidation catalysis Hairpins carrying an electroactive molecule (beacons) Electroactive intercalators

3. 3 Capacitance can also be used to detect specific label-free sequences, as hybridization induces charge effects, altering the dielectric properties of the biolayer in-between the metallic fingers  capacitance change. Enzymatic labels. When hybridization occurred, the enzyme is brought close to the surface, giving rise to an amperometric current proportional to the amount of hybridized strands. As no current was obtained from the enzyme-modified oligonucleotide in solution (because of the large distance between the enzyme and the redox polymer), no washing step was needed and the format provided measurements in real time  signal-on. In another format, the enzyme produce smthg that impede electron transfer on the surface  signal-off. Nature Biotechnology 21, 1192 - 1199 (2003)

4. 4 Other DNA beacons Template-enhanced hybridization. A template (the target) provokes the formation of a ternary complex (“Y” junction structure) with two other partially complementary sequences. This method is particularly sensitive to mismatches. The concept of template-enhanced hybridization, using Ru complex as reporter. Reprinted from Zhang and Chen et al. 2009. RuHex: ruthenium hexamine (electroactive) The Ru complex can be reduced or oxidized reversibly on the electrode where the DNA is immobilized.  The current is proportional to the qty of Ru, therefore to the qty of DNA.

5. 5 CNTs are promising for their high surface-to-volume ratio and electron transfer (ET) properties. Several architecture have been described. Co-immobilization of a redox indicator and a DNA probe on vertically-aligned SWCNTs. Before hybridization, the redox indicator forms hydrogen bonds with the DNA probe whereas these bonds are broken upon hybridization, which causes an increase in the indicator’s electroactivity. Use of CNTs or graphene Graphene based DNA electrochemical sensors were reviewed recently (Gao et al. 2014). Example: original approach by De Souza et al. 2014, described reduced GO modified by carboxylic acid- functionalized cobalt porphyrin. The carboxylic groups were used to covalently graft ODN probes. It was shown that the catalytic oxygen electroreduction on cobalt porphyrin is influenced by hybridization and is favorable compared to non-catalytic amperometric system. Apart from oxygen, no added reagent is nec- essary. A LoD in the pM range was reached. miRNA detection is based on an increase in the catalytic oxygen reduction current on cobalt porphyrin immobilized on reduced graphene oxide.

6. 6 g. Optical methods The oligonucleotide is labeled with an indicator dye: A fluorophore An enzyme In colorimetric DNA sensors, an enzyme is linked to the ODN. The enzyme converts its substrate into a colored product:  absorbance change is proportional to the hybridization efficiency. With label In fluorimetric DNA sensors, a fluorescent dye is linked to the ODN.  Fluorescence proportional to the hybridization efficiency. Fluorescence techniques also use molecular beacons. Molecular beacons are hairpin-forming probes with a fluorescent moiety at one end and a quenching moiety at the other. When they are not hybridized with the target, the hairpin or stem-loop structure maintains the termini at quenching distance. In the presence of the target the hairpin opens because the loop portion of the molecule is complementary to the target, the two ends are separated, resulting in a fluorescent signal. This also works when the ODN probe is immobilized on a surface.

7. 7 Without label Surface plasmon resonance (SPR) does not require any label and allows real-time measurement. The refractive index of the sensing layer changes depending on the amount of DNA in proximity with the surface. ‘‘SPR imaging’’ (SPRi), based on the same SPR principle with the exception that the metal surface is imaged on a CCD camera via an imaging lens, can be applied to array formats. The SPR system is particularly useful for the determination of binding and dissociation kinetics.

8. 8 Without label Surface plasmon resonance (SPR) does not require any label and allows real-time measurement. The refractive index of the sensing layer changes depending on the amount of DNA in proximity with the surface. ‘‘SPR imaging’’ (SPRi), based on the same SPR principle with the exception that the metal surface is imaged on a CCD camera via an imaging lens, can be applied to array formats. The SPR system is particularly useful for the determination of binding and dissociation kinetics.

9. 9 h. Transistors (FET) Ion-sensitive field-effect transistor (ISFET) (Bergveld 2003), charge-modulated FET (CM-FET), organic electrochemical transistor (OECT), organic FET (OFET) and electrolytic-gated organic FET (EGOFET) are the most reported architectures used for biodetection The circle indicates the interface involved in the biodetection FETs are sensitive to changes in surface charges they are suitable for detection of DNA. Electrochemical transistors In OECTs, source and drain are connected via a conducting polymer, separated from the gate by an electrolyte. The operating principle of OECTs relies on the doping and dedoping of the polymer, which results in modification of its conductivity. OECTs were described for DNA sensing, for example using PEDOT:PSS. Label-free DNA detection was obtained through the modulation of the surface potential of the gate electrode induced by target hybridization, with a LoD of 10 pM.

10. 10 Graphene-based FET devices able to detect DNA hybridization were described with LoD in the fM range, attributed to electronic doping induced by target DNA. Left. Transfer characteristics for the graphene transistors before adding DNA, after immobilization with probe DNA, and after reaction with complementary or (Right) one-base mismatched DNA molecules with the concentration ranging from 0.01 to 500 nM.

11. 11 Ion-sensitive field effect transistors ISFET were investigated for DNA sensing (first successful attempt: 1997). Left. Si 3 N 4 -based ISFET using a ssDNA layer on top of the dielectric. A reference electrode is necessary.. Middle. DNA immobilization and hybridization lead to a threshold voltage (V T ) shift. Right. Time-dependent channel current of an OECT measured after applying different gate voltages. V DS = −0.1 V.

12. 12 Electrolyte-gated field-effect transistors (EGOFET) EGOFET works at very low potential (> 1V). An EGOFET DNA sensor based on P3HT and poly[3-(5- carboxypentyl)thiophene-2,5-diyl] (P3PT-COOH) was described in 2012. Clear changes in the output characteristic of the device were observed upon DNA immobilization and after DNA hybridization. Results pointed out the importance of the Debye length that can screen negative DNA charges and impede transduction. Left. P3HT and P3PT–COOH solutions were spin-coated between source and drain. For measurements, a water droplet was deposited onto the semiconductor and a platinum wire dipped into the droplet was used as the gate. Right. Transfer curves for hybridization with a complementary target, and without target.

13. 13 Remote gate Using 6,13-bis(triisopropylsilylethynyl) (TIPS) pentacene as the organic semiconductor and an original architecture with a remote sensing area. The device shows a sub-nanomolar detection limit and unprecedented selectivity with respect to SNPs. Left. Structure and detection mechanism. Right. Transfer characteristics. [ODN target ] = 100 n M. Inserts: fluorescence images of the sensing surface (the target ODN were labeled with a fluorescent dye).

14. 14 Nanowires (other type of transistor) NW-based FET sensors are promising because of their high surface-to-volume ratio makes them sensitive to minor surface perturbations (e.g., binding of biomolecules). Publications reviewing nanowire-based FETs for DNA sensing are available, considering classical DNA probes or neutral probes (PNA or other neutral DNA derivatives). The sensing mechanism is based on changes in charge density at the NWs surface after hybridization. Left. DNA hybridization measurements at SiNW functionalized with ODN probes. (A) Real-time conductance response to 60 fM DNA. The inset shows a SEM image of a typical SiNW device with source (S) and a drain (D) indicated; scale bar is 1 μm. Middle. Conductance modified with DNA probes, where the arrow indicates the addition of 25 pM complementary DNA for (b) for p-type SiNW. Right. For an n-type SiNW.

15. 15 Carbon nanotubes (other type of transistor) CNTs were implemented in FETs (CNTFETs). Due to the high surface-to-volume ratio, the electrical resistance of nanotubes is extremely sensitive to changes in charge density at their surface. Hybridization on DNA- functionalized CNTs hybridization on DNA-functionalized CNTs can therefore be detected by measuring the source-drain current through nanotubes, function of the gate (or reference) potential. There are two classical types of device. The first design uses a single carbon nanotube to act as an electron channel between the source and the drain electrodes whereas the second type involves a network of CNT as a collective channel. DNA hybridization using CNTFETs was reported using PNA, with LoD as low as 6.8 fM (Maehashi et al. 2004). The transduction mechanism is still under discussion but the role of charges is confirmed. Left AFM picture of a CNTFET with a semiconducting SWNT contacted by source (S) and drain (D), above a SiO 2 insulating layer. Middle. SEM image of a random array of CNT. Right. Typical CNTFET transfer characteristics (A: threshold voltage; B: transconductance; C: maximum conductance; D: modulation).

16. 16 Break junctions and nanogaps Nanogaps were described by Hashioka et al, 2004 for DNA detection, generally based on gold electrodes (used to graft DNA probes) separated by a 40 nm gap. The current is modulated by the presence of DNA strands in between the two electrodes. Interestingly, arrays of multi-junctions were also achieved by construction of a Au-NPs film above a classical interdigitated electrode. SWNTs were also used for the same goal. Upon hybridization, dsDNA act as conducting channels and increase current. Left and Center. Gold and titane electrodes separated by a 40 nm gap, used to electrically detect DNA strands. Right. Schematized top view of a AuNPs film with bridging dsDNA.

17. 17 Nanopores and nanochannels Nanopores or nanochannels present a diameter similar to ssDNA or dsDNA. Nanopores are generally protein channels embedded in lipid bilayer or apertures in synthetic membranes fabricated in a variety of materials (plastic, glass, or silicon). Upon application of potential difference across the membrane, an ionic current goes through the nanopore, but is blocked during translocation of a DNA strand through the pore. Initially, biological nanopores were used, e.g.  -haemolysin (  -HL) which presents a suitable diameter. The DNA sequences can even be identified by this way. Identification of nucleobases is also possible using conducting nanogaps inside nanochannels, able to measure the electrical conductivity perpendicular to the channel. Left: 1- An α-HL nanopore, and 2- a typical solid-state nanopore in a silicon nitride membrane. Middle: Ionic current recorded as a function of time, during a translocation process. A, direct translocation; B, DNA is captured then exits from the entry; C, DNA is captured in the vestibule before it translocates; D, DNA enters the barrel then retracts from the entrance; E, DNA explores the vestibule, threads into the barrel, retracts back into the vestibule and exits from the entrance. Right: Schematic principle of a nanochannel.

18. 18 4. Amplification methods (general) Sensitivity is the one the factor of merit. For example: for a viral infection, the amount of DNA that has to be detected is at the attomolar (10 - 18 M) level. Two possibilities Amplify the signal Increase the amount of sample (or both at the same time) a. PCR Technique most commonly used. During PCR, the amount of DNA is exponentially amplified by repetitive cycles. PCR is also used to label the target with antigens, fluorophores and labile groups, when the detection technique used to measure hybridization is not label-free. Today, PCR is integrated into the detection chip.

19. 19 4. Amplification methods (general) Sensitivity is the one the factor of merit. For example: for a viral infection, the amount of DNA that has to be detected is at the attomolar (10 - 18 M) level. Two possibilities Amplify the signal Increase the amount of sample (or both at the same time) a. PCR Technique most commonly used. During PCR, the amount of DNA is exponentially amplified by repetitive cycles. PCR is also used to label the target with antigens, fluorophores and labile groups, when the detection technique used to measure hybridization is not label-free. Today, PCR is integrated into the detection chip.

20. 20 4. Amplification methods (general) Sensitivity is the one the factor of merit. For example: for a viral infection, the amount of DNA that has to be detected is at the attomolar (10 - 18 M) level. Two possibilities Amplify the signal Increase the amount of sample (or both at the same time) a. PCR Technique most commonly used. During PCR, the amount of DNA is exponentially amplified by repetitive cycles. PCR is also used to label the target with antigens, fluorophores and labile groups, when the detection technique used to measure hybridization is not label-free. Today, PCR is integrated into the detection chip.

21. 21 b. RCA It’s a hybridization signal-amplification technique. It uses padlock probes (circularizing oligonucleotides). Once the probe is immobilized, the complementary target sequence is added and hybridizes. Afterwards, a second probe is added, hybridizes with part of the target and is linked to the first probe by a DNA ligase. Then, the target is removed and the padlock probe is added. In the presence of a strand-displacing DNA polymerase, the primer is extended and, after one complete revolution of the circularized probe, the primer is itself displaced at its 5’ terminus. Continued polymerization and displacement generates a single-stranded concatameric (repetitive) DNA copy of the original probe. To detect hybridization, fluorescent complementary tags are added and hybridize at the multiple repeated sites in the elongated DNA sequence.

22. 22 Branched DNA (bDNA) is a hyperpolymeric DNA. The use of amplifier bDNA typically requires two hybridizations : first between the immobilized probe and the target, and second between the target and some preamplifiers; amplifier bDNA structures are then hybridized to the preamplifiers. The bDNA has multiple single-stranded arms either available for consecutive conjugation reaction with labels or for further hybridization with labeled sequences. The use of bDNA amplifes the hybridization signal and lowers the LOD. c. Branched DNA Capture Label Fluorescen ce labelling

23. 23 Special case of dendritic amplification of DNA: based on ODN-functionalized gold NPs used in piezoelectric techniques. These NPs contain ODN fragments that hybridize with the target once it has already hybridized with the immobilized probe, thus amplifying the frequency change. Further amplification can be achieved using gold NPs capable of hybridizing with the ODN fragments of the first one. By choosing the appropriate size of NPs, a LOD of 10 -14 M was achieved. J. Mater. Chem., 2006,16, 3997-4021 Nanoscale, 2013, 5 (20), 9503–9510 d. Dendritic NPs Amplif. can also be performed using Au NPs in a sandwich config. Ag° is grown after hybridization to make the assembly heavier -> mass change A labeled primary probe can be hybridized first. Displacement (by the target) removes the electrochemical label  signal-off

24. 24 2 nd evaluation 1- Structure of nucleic acids a.DNA b.RNA c.PNA d.Nucleobases 2- Making DNA-modified surfaces (chips) a.Membranes b.Polymers c.Glass and silicon d.Contact and non-contact printings 3- Transduction techniques a.Optical b.Electrochemical c.Transistors d.Nanowires

25. 25 5. DNA probes for non-DNA targets: Aptamers 94 -Proteins S. Xiao et al., Anal. Chem., 9736–9742, 2010 -Amino acids C. Lozuponz et al., RNA, 1315–1322, 2003 -Insecticides H. Jiang, et al., J. Agric. Food Chem., 1582–1586, 2011 -Drugs P.C. Anderson et al., J. Am. Chem. Soc., 5290-5291, 2005 Advantages: High affinity (nM <K d < µM) Selectivity / Enantioselectivity Small molecules / toxins Easily functionalized Drawbacks: Sensitive to nucleases Affinity < than Antibodies < < Protein binding Small organic molecule binding < < a. Generalities

26. 26 Nucleic acid aptamers are single-stranded DNA or RNA sequences which bind selectively to a target molecule (mostly organic molecules) through folding into a complex three- dimensional structure. A.D. Ellington, J.W. Szostak. Nature 1990, 346, 818–822. Target – Aptamer Interactions includes : structure compatibility, stacking of aromatic rings, electrostatic and van der Waals interactions, hydrogen bonding. The most common aptamer : vs. thrombin (a protein). It has been shown that the thrombin aptamer (TBA) takes a special conformation, a G- quadruplex induced by K + before thrombin can be specifically bound. GGG-TTA-GGG-TTA-GGG-TTA-GGG K+K+ K+K+ thrombin

27. 27 Nucleic acid aptamers are single-stranded DNA or RNA sequences which bind selectively to a target molecule (mostly organic molecules) through folding into a complex three- dimensional structure. A.D. Ellington, J.W. Szostak. Nature 1990, 346, 818–822. Target – Aptamer Interactions includes : structure compatibility, stacking of aromatic rings, electrostatic and van der Waals interactions, hydrogen bonding. The most common aptamer : vs. thrombin (a protein). It has been shown that the thrombin aptamer (TBA) takes a special conformation, a G- quadruplex induced by K + before thrombin can be specifically bound. GGG-TTA-GGG-TTA-GGG-TTA-GGG K+K+ K+K+ thrombin < < Identified Structures G-Quadruplex (Thrombine, ATP…) Hairpin (Theophylline, AMP…) Multiple Stems (Cocaine…)

28. 28 < < < < 3’-ATTAAAGCTCGCCATCAAATAGC-5’ On the same basis than for immunosensors, aptamers may be immobilized on a redox polymer-modified electrode Biosens. Bioelectron., 2015 Example : protein detection (PSA) [PSA]

29. 29 < < < < The Aptamer Handbook: Functional Oligonucleotides and Their Applications, S. Klussmann (Editor), Wiley, 2006 Problem: aptamers have better affinity to big proteins than small molecules LOW MOLECULAR WEIGTH TARGET KDKD TARGET 50-70 µM 23 µM 1 µM 6-13 µM PROTEINS KDKD TARGET  -Thrombin Ig E HIV-1 Tat Interferon-  VEGF L-Selectin 25 nM 9 nM 0.12 nM 3 nM 0.15 nM 1.8 nM Arginine L-Tyrosine Malachite ATP Cocaïne

30. 30 Some examples Sensors 2013, 13(10), 13928-13948 The company NeoVentures Biotechnology Inc. (http://www.neoventures.ca) has successfully commercialized the first aptamer-based diagnostic platform for analysis of several toxins in water.http://www.neoventures.ca

31. 31 Hemin-conjugated DNA hairpin and its structural change upon recognition of Ochratoxin A, which forms an active G- quadruplex. b. Enzyme mimicking The label provides a flux of detectable species amplification. For example: Ochratoxin A (OTA). The OTA aptamer binds OTA, then the detection sequence is freed and forms a G-quartet Hemin becomes active. Hemin oxidize ABTS into its oxidized form (coloured product)  optical detection

32. 32 c. Target recycling Target amplification cannot be applied in the case of aptamers as the target is not a nucleotide sequence. Below, an example of FRET (fluorescent transduction system seen before) but coupled to target recycling: In the presence of the target (here, VEGF, a growth hormone) and a short template sequence, the aptamer conformation switches to a G-quartet in the presence of the target, activating one of the sequence positions toward exonuclease (Exo III). Sequence digestion with exonuclease releases both the quencher, leading to an intense fluorescence emission, and also the target, which may then be bound again. FRET analysis of VEGF165 by CdSe/ZnS QDs modified with the anti-VEGF aptamer. Principle of the Exo III-aided ATP recycling strategy and its associated homogeneous electrochemical detection. Exonuclease digesting a ferrocene-terminated aptamer only when the latter forms a complex with its target, here ATP. The redox probe is released. The difference in diffusion coefficient of free ferrocene and ferrocene grafted to the DNA allows to correlate the current value to the amount of target.

33. 33 d. Enzyme amplification (not as label) Schematic Representation of Homogeneous Aptamer and Nicking Enzyme Assisted Fluorescence Signal Amplification (NEFSA) Assay for Protein. A nicking enzyme can do the same job. The enzyme binding site is located on a complementary strand that hybridizes to the aptamer detection sequence that is only available after target/aptamer complex formation. When the detection sequence is released, fluorescence occurs, and another functionalized sequence can hybridize to the aptamer. At each cycle, the fluorescence signal increases. RCA elongates a detection sequence that is available only after target binding. The general approach is based on an aptamer conformation change upon target binding that triggers cyclisation of the detection sequence and then RCA. The elongated sequence serves as a support for sensing, either using complementary DNA sequences alone or fluorophore label ones. e. RCA Schematic of deoxyribozyme-mediated RCA. The aptazyme labelled with biotin at its 5‘ end is immobilized on a streptavidin-coated glass slide. The aptazyme is activated by ATP and ligates a probe. RCA is initiated from the 3‘ end of the aptazyme, and the elongated aptazyme is visualized using fluorescent oligonucleotide probes labelled with Cy3.

34. 34 b. Chiral detection (electrochemical approach) Enantioselective molecular recognition of trace amounts of L- or D-tyrosinamide. The chirality of the phosphate backbone is an intrinsic structural property of nucleotides that remains underutilized. The natural state of DNA is the D-enantiomer due to sugar chirality. DNA is readily synthesized as L-enantiomer with conventional technology using L-sugars nucleobases. It is now possible to synthesize enantioselective D-aptamers against L-Targets and L-aptamers against D- target. Most current analytical approaches developed in the context of enantiomeric excess (e.e.) measurements are mainly separation methods. There is plenty of room for sensors ! One promising principle: to discriminate between the complexed and free fractions of the target on the basis of their transport properties in solution. Indeed, upon binding, the target solution has a molecular weight that artificially increases, decreasing its solution transport characteristics. Such a binding event can then be followed by measuring a current, provided that the target is electroactive. This method is illustrated here within a competitive exchange scheme using a target functionalized with a redox probe.

35. 35 B. Protein biochips 1. A few definitions a. Antigens (Ag) Ag: biological or synthetic molecule recognized by antibodies. So, an antigen is an immunogenic organic molecule. Ag are generally proteins or polysaccharides. Ag fragments (named haptens) can also complex antibodies (Ab). For proteinic Ag, epitopes are the Ag part that is recognized by the Ab. One single Ag can be made of several epitopes. It exists : Sequence epitopes, corresponding to an AA sequence (generally short sequences, a few AA) Conformational epitopes, longer in AA (from 10 to more AA) b. Antibodies (Ab) An Ab is a macroprotein used by living organism to detect and eliminate (digest) Ag. Ab (IgG) are formed by 4 polypeptidic chains (of 150 kD): 2 heavy chains (H, 50 kD, in violet) and 2 light chains (L, 25 kD, in green) These chains make a Y structure and are made of domains of ca. 110 AA. Each light chain is made of a constant domain and a variable domain; heavy chains are made of a variable domain and 3 to 4 constant fragments.

36. 36 Specific enzymatic clivage allows to isolate various fragments : F c fragment. Not interesting here (no recognition properties). F v fragment. It is made of variable regions VL (various light) & VH (various heavy). It is specific of the Ag. F ab. It is made of VL+CL (constant light) & a part of the VH chain. F (ab')2. It corresponds to the association of 2 F ab fragments link together by a small part of the CH (constant heavy chain). It complexes the Ag. Monoclonal Ab are Ab that recognize only one epitope on one given Ag. Monoclonal Ab are the most frequently used in biochips (because of their specificity) (by contrast to polyclonal Ab). c. Monoclonal antibodies: How to downsize antibodies ? F (ab')2 is the most used in biochips. It is smaller then the whole Ab and keep an acceptable specificity. However, smaller probes will be needed in near future for biochips, to increase surface densities & sensitivities  only VL chains ?? Why antibodies are too big ? Because high surface area of probe needs small probes !

37. 37 … Example

38. 38 … Example

39. 39 … Example

40. 40 e. Peptide They are macromolecules, made of a series of polypeptides. There are 20 classical AA (those coded by m RNA). All have the same basic formula, with various -R functions. AA are linked by -NH-CO- (amide) in between the carboxylic acid of one AA and the primary amine of the other AA. It is a peptide bond. Peptide bonds : A peptide is a chain of less than 50 AA linked by peptide bonds. They are shorter than proteins. These polypeptides are made of a series of amino acids (AA). Polypeptide : A polypeptide is a chain of more than 50 AA. Example : Lys-Val-Phe-Gly-Arg-Cys-... It is its primary structure Because of sequences composed of 20 AA, these proteins carry an information that is very rich (20 x for x AAs) (so, ++ rich than DNA, 4 x for x bases). d. Proteins

41. 41 3. Peptide biochips Making the array: same methods as for DNA Peptides can substitute for proteins in biological analysis. Peptides with specific sequences can provide high affinity to particular analytes, and be obtained by screening and optimization of artificial peptide libraries. Peptides have shown further advantages: high stability, standard synthetic protocol, easy modification and large chemical versatility. For example, peptides with short chains of amino acids generally have better chemical and conformational stability than proteins. K d is the lowest for peptide/protein interactions Peptides are the corresponding antigens (haptens) Peptide/cells (e.g. bacteria)

42. 42 The 22 natural amino acids

43. 43 Proteins functionalities are directly related to their spatial arrangement (if the structure is changed, the functionnality is changed as well : denaturation). The functional group -R governs peptide conformation (secondary structure) and protein folding (alpha helix, beta sheet) (tertiary and quaternary structures).

44. 44 What governs protein conformation (spatial arrangement) (from strong to weak) : + Disulfide bridge (R-S-S-R) + Ionic interactions + H bonds + Van der Waals. Examples of protein folding : 3 foldings :  Beta sheets (planar)  Alpha helix (helicoidal)  Random

45. 45

46. 46 a. Peptides as recognition elements Peptides are ideal candidates to replace proteins as receptor (biorecognition element) in biosensors. Artificial peptides can be obtained through standard solid-phase synthesis to provide a specific sequence or screening a library of peptides. These peptide-based molecular biosensors have been developed for convenient, fast detection of various analytes including proteins, antibodies, DNA and metallic ions. Peptide-based protein sensors Most peptide-based protein sensors rely on fluorophores. Transduction mostly relies on fluorescent resonance energy transfer (FRET). Typical transduction schemes are : (A) Peptide probe with environment-sensitive fluorophores as signal markers to respond to the target protein. Upon recognition, the binding event induces a change in the spectral properties of the fluorophore. The emission of fluorophores is usually enhanced and blue shifted after affinity/recognition in a polar solvent.

47. 47 (C) FRET/probe–quencher pair peptide sensor. The distance between the donor–acceptor/probe– quencher was changed by binding of the peptide to the protein. In probe– quencher pair protein-sensors, fluorophores and quencher units are attached at various residues of the peptide sequence. Affinity between a peptide sequence and target protein influences the distance between the fluorophore and quencher, resulting in a change in the fluorescence emission. (B) excimer-pair peptide sensor. The peptide binds to the target protein, changing the spectral property of the fluorophores; Excimer-type protein molecular biosensors are composed of a peptide and two identical fluorophores, which are attached to the opposite ends of the peptide. In the presence of the target analyte, the interaction between the analyte and peptide forces the fluorophores to seperate, resulting in a shift of the emission peak of the fluorophores to that of the monomer.

48. 48 Electrochemical peptide-based protein sensors. They are based on bio-conjugation strategy. For example, by conjugating an electrochemical marker with a peptide sequence that has a high affinity for the target protein, and immobilizing the peptide bioconjugate on the electrode of the sensor, the binding of the analyte with the peptide sequence can considerably change the electron transfer from the electrochemical marker to the electrode, which would be detected by the sensor. Schematic diagram of an electrochemical peptide-based protein sensor. The electron transfer (ET) from electronic marker to the electrode is influenced by the binding/affinity between the peptide and the target protein.

49. 49 Peptide-based metallic ion sensors Peptides with special sequences can also interact with metallic ions. The capability of peptides to capture metallic ions can be utilized to construct peptide-based metallic ion sensors by conjugating the special peptide sequence with signal markers. Most peptide-based ion sensors rely on fluorophores. The peptide sequences used in metallic ion sensors can be obtained by optimizing peptide libraries or using special protein fragments. A wide variety of metallic ions, including Zn(II), Cu(II), Hg(II), Cd(II), Ca(II), Mg(II), Na(I), K(I), Mn(II), Co(II), Al(III), Ni(II), Pb(II) and Ag(I), can be detected by such metallic ion sensors. Same transduction architecture as for DNA or proteins

50. 50 However, there is several example of electrochemical peptide-based metallic ion sensors For example, transduction can be base on differential pulse adsorptive stripping voltammetry (DPASV) (A) GSH, γ-Glu–Cys and Cys–Gly electrografting; (B) Measured vs. expected concentrations for Cd 2+. The LoD is below 20 ppb. (A) (B) Same transduction architecture as for proteins ASV : the ion is first pre-concentrated on the surface, by the peptide probes. Then, a strong reduction potential is imposed, so that the ions are reduced into M° (Adsorption). Then, stripping consists in fast re-oxidation into M 2+, which produces a high current.

51. 51 Biosensors Bioelectronics 2013, 48, 263–269 Short peptides may be used as probes, for ions detection < < Peptide : TNTLSNN on poly(thiophene) K d = 3.3×10 −11 M to Pb 2+ at pH 4 Peptide Aptamers Gold deposited by dynamic hydrogen bubble template method (10 mM HAuCl 4 under −3 or −5 V during 10 to 30 s Thr-Asn-Thr-Leu-Ser-Asn-Asn Pb 2+ 1 st example : Pb 2+ detection

52. 52 Biosensors Bioelectronics 2013, 48, 263–269 Short peptides may be used as probes, for ions detection < < Peptide : TNTLSNN on poly(thiophene) K d = 3.3×10 −11 M to Pb 2+ at pH 4 Peptide Aptamers Gold deposited by dynamic hydrogen bubble template method (10 mM HAuCl 4 under −3 or −5 V during 10 to 30 s Thr-Asn-Thr-Leu-Ser-Asn-Asn Pb 2+ 1 st example : Pb 2+ detection

53. 53 < < Electroanalysis 2013, 25, 1461–1471 Gooding et al. Gly-Gly-His Cu 2+ < < 3 rd example : Cu 2+ detection

54. 54 < < Electroanalysis 2013, 25, 1461–1471 Gooding et al. Gly-Gly-His Cu 2+ < < 3 rd example : Cu 2+ detection

55. 55 Proteases can hydrolysis peptides (proteolysis). Development of protease sensors is significant for human health. Protease-sensitive peptide sequences are necessary Signaling mechanism using a protease-cleavable peptide strategy. Donor– acceptor/probe– quencher pairs are attached on a peptide in close proximity, and cleavage of the peptide by the protease releases the donor/probe, resulting in an increased fluorescence emission intensity. Scheme of QD-based protease activity sensors. QD fluorescence is quenched by the attached quencher on the surface of QD connected by FRET. QD fluorescence is recovered after the protease cleaves the peptide linkers. As for other protein sensors, fluorescence sensors uses generally FRET Protease/protease activity sensor Electrochemical sensors are also described: Example: electrochemical assay for PSA. (A) Self- assembly of Fc-functionalized peptide probes on a gold electrode surface and electrochemical readout of redox activity of Fc. The PSA cleaves the peptide sequence to remove the Fc segments from the gold electrode surface (B), and reduces electrochemical readout of redox activity of Fc (C).

56. 56 3 rd evaluation 1- DNA probes for non-DNA targets a.Targets? b.Transduction 2- Peptides and proteins a.Structures b.Transduction

57. 57 Sugars (carbon hydrates) are made of carbon, oxygen and hydrogen. They are monosaccharides, disaccharides, etc. as a function of the number of units. Oligosaccharides are made of several carbohydrates (longer chains are polysaccharides). Even longer entities (associated with proteins or lipides) are glycoproteins or glycolipides, respectively. (a) Example of a monosaccharide, D-glucose. (b) Example of a disaccharide made of two monomers: mannose and glucose. C. Oligosaccharide biochips Oligosaccharides can recognize proteins, antibodies, cells, bacteria… DNA probes are complex. Peptides are more. Oligosaccharides are even more !

58. 58 1. Functionnalization of surfaces with carbohydrates a. Immobilization through the biotin / streptavidin system(historical approach) a) Structure of biotin ; b) Avidin-biotin interaction Sugar array, obtained by immobilisation through biotin-modified oligosaccharides on a streptavidin- modified 96-wells plate. b. Covalent grafting: dextran Method: Dextran is immobilized on an hydroxylated surface hydroxylée, then carboxylated. Dextran: branched polysaccharide made of many glucose molecules) (3 to 2000 kDa) High hydrophylic surface Oligosacharides are then immobilized covalently NHS : N-hydroxysuccinimide ; EDC : N-ethyl-N’-(3-diethylaminopropyl)-carbodiimide ; NH 2 -NH 2 : hydrazine ; Sulfo-MBS : m-maleimidobenzoyl- N-hydroxy-sulfo-succinimide ester.

59. c. Other covalent graftings Maleimides on thiols 59 Self-assembled monolayers (SAMS) Maleimide can react on one side on thiolated surface and on the other side with the sugar (works with monosaccharide & oligosaccharide). Oligosaccharides can be reacted on pre-formed SAMs through diene/quinone coupling (Diels-Alder)

60. 60 Phospholipides (on SAMs) Mono- or oligosaccharide can be first react on phospholipids, then assembled on a first layer of alky chain self-assembly through hydrophobic interactions. Disulfide bridges (on SAMs) MUAM: 11-mercaptoundecylamine Mono- or oligosaccharide can be first modified with a thiol, then coupled to a SAMs bearing an activated disulfide bridge (thiopyridone).

61. 61 d. Non-covalent grafting The oligosaccharide is first modified with a phospholipid. That phospholipid sticks on hydrophobic surfaces. Functionalization with a gycolipide Grafting on polystyrene plates An oligosaccharide functionalized by a long aliphatic chain (13 to 21 carbons) sticks sufficiently well (through hydrophobic interactions) on polystyrene, even after several aqueous washings. Grafting on glass surfaces bearing phosphane groups

62. 62 2. Methods of transduction Currently, the detection methods used for glycochips are quite limited and already known for their use in the DNA microarrays technology. Detection with label Detection without label Direct detection Indirect detection substrate product Substrate (material) Fluorescence Enzymatic detection Biacore, SPR Light sourceDetector

63. 63 This technique is efficient but cannot provide quantitative and/or kinetic informations. a. Direct detection Fluorescence BIAcore principle. Left: device; Right: reflectivity changes Principle of surface plasmons SPR

64. 64 Example of SPR detection 1st step: surface functionalization Structure of pyrrole-lactosyl (1) and pyrrole-39-sialyllactosyl (2). Analyst, 2008, 133, 206–212 The oligosaccharide is coupled to an electropolymerizable monomer (here, pyrrole). Electrooxidation of this molecule gives an oligosaccharide-modified surface, with adjustable surface density. 2nd step: SPR measurements Specific interactions on polypyrrole-lactosyl surface (5 nmol cm -2 ) (PNA is the target lectin) [lectins are proteins known to bind carbohydrates] 3rd step: Calibration & quantification Calibration curves of polypyrrole-lactosyl film with PNA lectin for different polymer thicknesses: 15 nm, 33 nm, 40 nm, 56 nm. K d (PNA) = 4.45  10 -7 M thickness

65. 65 b. Indirect detection Enzymatic amplification No example (yet) in the literature !

66. 66 Electrochemical detection The oligosaccharide is coupled to an electropolymerizable monomer (here, hydroxynaphthoquinone). Quinone is electroactive. Displacement of the Ab should change the ionic current at the interface, therefore change the faradaic current. Competitive immuno-exchange In this approach (click chemistry), any oligosaccharide can be coupled to the monomer  versatility. It can also allow to vary the linker (length and chemical nature [hydrophobic, hydrophilic]), to limit non-specific adsoprtion and enhance probe accessibility.

67. 67 Transistors Anal. Chem. 2013, 85, 5641−5644 A number a oligosaccharide-modified transistors have been described in the literature. Below, is described a recent approach featuring a Si-based FET The ref. electrode is the true gate. The Si0 2 isolator is used as “second gate”. Its surface is first functionalized by aminooxy terminated trialkoxysilane, which is then able to react on unmodified oligosaccharides (opening the sugar ring). 6’-sialyllatose can recognize hemagglutinin for the influenza virus. This method allows to “blott” (i.e. make spots) any oligosaccharides on aminooxy-modified silicon. 6’-sialyllatose

68. 68 Vg – Id characteristics before and after the adsorption of HA on glycan-immobilized gate surface. The ∆Vg shifted in positive direction by 71 mV after the addition of 500 pM H5 HA. How does it work ? Hemagglutinin (HA) has an isoelectric point of 6.85. In PBS (pH 7.2), HA is therefore negatively charged (deprotonated). The adsorption of HA molecules turns the surface charge to negative, which cause the ∆V g (a more positive potential at the gate compensates the negative potential brought by HA). Quantitative determination of HAs using glycan- immobilized FETs. Detection of human influenza virus H1 HA and avian H5 HA for Sia-α-2,3′Lac-functionalized FET (500 zM−500 pM). Error bars represent the standard deviation (N = 3−4). Note the extremely low detection limit.

69. 69 EXAM.