“Sensitive teeth are caused by the movement of fluid in tiny holes in dentine called microtubules, which puts pressure on the nerve below,” says Nigel Hughes, one of the researchers at Unilever. “This may occur when the dentine becomes exposed by receding gums or perhaps after surgery. Teeth are particularly sensitive to cold drinks and ice cream because this causes the fluid to contract, moving the nerve.”

There are two approaches to treating the problem. The first is to block the microtubules by adding chemicals such as fluoride to toothpaste to encourage calcification of the tooth. The second is to add desensitisers such as potassium salts, which diffuse into the tooth via the microtubules and depolarise the nerve. The flow rate of fluid through the microtubules plays an important role in determining how effective toothpaste additives are.

The SECM was pioneered by the American electrochemist Allen Bard and is a modified version of the scanning tunnelling microscope. The SECM uses a flat-faced probe a few micrometres across, which forms one electrode in an electrochemical cell. The other is a reference electrode.

The sample is bathed in an electrolyte solution called a mediator, which exchanges electrons with the electrodes, thus generating an electric current. The size of the current depends on how fast the mediator can diffuse to the probe. Away from a surface, it can diffuse in all directions, but close to a surface diffusion becomes restricted and the current falls (see Diagram). So as the probe scans across the surface of a sample the current falls as the surface rises to meet the probe. The changing current can be translated into a profile of the surface.

“Although the resolution of a few micrometres is much lower than that achieved with the scanning tunnelling microscope, what we have is more than just a device for producing pictures of the topography of surfaces,” says Patrick Unwin, one of the chemists at Warwick. “What we see is really a measure of how fast the mediator is supplied to the electrode and anything that changes this can be imaged.” Electrochemists have previously used SECMs to image corrosion and dissolving crystals. But Unwin’s group is the first to apply the SECM to measuring fluid flow.

In earlier experiments the researchers showed that they could produce flow profiles for very small jets, and they have now applied this approach to dentine. First, they glued a slice of dentine over the end of a tube containing a mediator solution and scanned the slice to produce an image of the surface. They then pressurised the fluid to force it through the microtubules and scanned the surface again. This time, as the probe crossed an open tubule, the supply of mediator increased, leading to a sharp rise in current.

“All previous methods measure the bulk flow rate and produce an average for an individual tubule after the number of holes has been estimated,” says Hughes. “For the first time we do not have to make any assumptions but can measure the flow rate on the local scale.”

The team found that many microtubules were blocked, registering no flow when pressure was applied. And they discovered that the flow rates for unblocked holes were up to a thousand times higher than previous estimates (Faraday Transactions, vol 91, p 1407). “Because individual rates appear so high we may have to reassess models describing the inward diffusion of desensitisers,” says Hughes.

If, like Alice, we could step through a looking glass into a mirror-image world, what would it be like? More importantly, could we survive? Scientists have now made the first inroads in creating such a ‘through the looking glass’ world.

Stephen Kent and his colleagues at the Scripps Research Centre at La Jolla, California, have made the first active d-chiral enzyme. This enzyme, d-HIV 1 protease, is the mirror image of the naturally occurring l-HIV 1 protease, an enzyme from HIV which cleaves certain peptide chains (Science, vol 256, p 1445).

Chirality is the property which makes a left-hand glove different from a right-hand glove, even though one is an exact mirror-image of the other. Organic molecules in which four different chemical groups are attached to a central carbon atom are also chiral, as they too can be constructed in two forms, usually labelled d and l and known as enantiomers.

Nature is inherently chiral. Proteins are almost exclusively made up of l amino acids whereas carbohydrates are made up of d sugars (There is no significance to the labels being different in the two groups. It is merely an accidental result of the way they are assigned). The chirality of the building blocks results in chirality in the bulk structure, in the same way that oppositely shaped wedges go together to make left and right-handed spiral staircases.

Following this logic, Kent’s group made the d-enzyme from exclusively d-chiral amino acids. They did this using the Merrifield method of protein synthesis, which effectively builds up a protein an amino acid at a time. He used the traditional method of optical rotation to show that the products were also chiral. This relies on the fact that solutions of pure enantiomers rotate the plane of polarisation of light passing through them; opposite enantiomers rotate it in opposite directions. Further evidence that the products were chiral was that each enantiomer would cleave only a peptide chain of the same chirality and was only inhibited by inhibitors of the same chirality. Such specificity is typical, if not universal.

What use could such unnatural products be? Not all protein interactions are chiral. Just as our chiral hands can manipulate perfectly symmetrical objects – pens, bats, balls and so on – so the iron-sulphur protein l-rubredoxin, binds achiral metal ions. In a paper soon to be published in the Journal of the American Chemical Society, Laura Zawadzke and Jeremy Berg at the Johns Hopkins University in Maryland have made the d version of this protein, which is as good as its mirror image at binding metal ions.

Defects in the non-chiral interactions of an individual’s enzymes could, in theory, be treated with d-chiral proteins. Such drugs would presumably last longer, because the body’s natural mechanism for degrading proteins could not break them up. (For the same reason we would have difficulty living in our ‘looking glass’ world because our bodies would not be able to process the indigenous proteins and carbohydrates). Chiral enzymes could also be used to produce enantiomerically pure drugs.

But there is a more immediate practical application, as Zawadzke and Berg’s work has shown. Whereas most organic chemists take great pains to separate mixtures of enantiomers that they synthesise, there are times when ‘racemic’ mixtures, containing equal quantities of the d and l forms, are desirable. For example, crystals from a racemic mixture are ideal for studies of protein structure using X-rays. One way of finding the structure of a naturally occurring l protein might therefore be to mix it with its d enantiomer before performing X-ray diffraction analysis.

Life through the looking glass may not be all bad after all.

The photochemical equivalent of the electrical on-off switch, made using a mixture of dyes, has been successfully developed by Itamar Willner and colleagues from the Hebrew University of Jerusalem. The switch is turned on by ultraviolet light and off by blue light. Such a system could eventually be used to build a superfast photochemical computer.

A conventional computer uses circuits that are switched on and off by electric currents. An optical computer operated by the reactions of single molecules to light could be much smaller and faster. Although researchers do not know how to design such a computer, they know that photochemical reactions can mimic some of the functions of microprocessors.

In a system based on Willner’s new ‘switch’, electric currents are replaced by beams of light and the electronic components by a mixture of eosin Y, a biological stain, and a compound of phenylazobenzene with viologen. (Viologens include the weedkiller Paraquat.)

Eosin Y is a red pigment – in other words, it absorbs green light. But when it joins certain other molecules to form a so-called charge-transfer complex, it becomes transparent to green light (Angewandte Chemie, vol 31, p 1243).

Willner’s success has been to find a second light-sensitive molecule which has two forms – one of which combines with eosin Y, making it transparent to green light, and the other which will not combine, leaving the eosin Y opaque to green light. This second molecule is the viologen-phenylazobenzene. The critical part of this compound is the azo group of the phenylazobenzene, which contains two nitrogen atoms joined by a double bond (see Diagram). Phenylazobenzene comes in two forms or isomers. In the trans isomer the two benzene rings lie on opposite sides of the double bond. In the cis isomer they are on the same side.

When the trans isomer absorbs ultraviolet light, the double bond becomes temporarily weakened and the molecule can change into the cis isomer. If this cis isomer then absorbs blue light, it can revert back to the trans form.

The conversion from the trans to the cis isomer is associated with a change in the dipole moment – the overall distribution of charge – of the viologen-phenylazobenzene molecule. The cis molecule has a higher dipole moment than the trans form.

Eosin Y can associate only with the viologen-phenylazobenzene molecule with a high dipole moment – the cis form. So when ultraviolet light is shone on this compound, converting it into the cis form, the eosin Y and viologen associate and green light is transmitted.

Hiroshi Yoneyama and his colleagues at Osaka University projected an
image onto a thin film of polyaniline which contains particles of semiconducting
titanium dioxide. By electrically manipulating, or ‘switching’, the polymer,
they were able to erase the image and use the film again (Journal of the
Chemical Society, Chemical Communications, 1992, p 715).

Polyaniline belongs to a class of compounds known as organic conducting
polymers. The electrical conductivity of these polymers depends on their
oxidation state – that is, the number of positive or negative charges in
their polymer chains.

Most conducting polymers have to be oxidised and carry a net positive
charge in order to conduct a current. They can be switched to their nonconducting,
or reduced, state by injecting electrons into them to cancel the positive
charge. Because the colour of a substance depends on how the electrons in
its molecules are arranged, the polymer changes colour.

The ability of these polymers to change colour has led to the suggestion
that they could be used in electrochromic displays. These would be similar
to liquid crystal displays (LCDs) used in watches and calculators, but they
would have the advantage that many more colours would be available (LCDs
are nearly always black).

A television screen based on the electrochromic effect might consist
of an array of thousands of tiny electrodes. As the voltage of the signal
is scanned over them, the dots represented by each electrode would change
colour. By using a range of polymers, it would be possible to produce a
full-colour picture. Such a system would dispense with the cathode-ray tube
and the high voltages at which it operates.

Yoneyama and his colleagues replaced the electrodes with particles of
titanium dioxide embedded in the polymer. They made the polymer film by
passing a current through a solution of aniline, a derivative of benzene.
The film formed on the positive electrode, or anode. When light was shone
on the film, electrons in the semiconductor became excited and were absorbed
by the surrounding polymer, changing its colour from yellow to green.

Unfortunately, because polyaniline is a good conductor, the charge generated
in the semiconductor is quickly carried away from its source. So even if
the film is used to depict an image confined to one place, the colour change
spreads through the film.

Yoneyama’s trick was to use the polymer in a neutral rather than an
acidic solution. Polyaniline conducts only in acid because it needs hydrogen
ions (protons) to assist its conductivity. Yoneyama included methanol, a
proton donor.

When the semiconductor particles absorbed light, this created excited
electrons and positive holes. Electrons were consumed by the polyaniline,
while the holes were consumed by the methanol, releasing hydrogen ions,
which made the solution acidic close to the place where the light was absorbed.
This allowed the polymer to be reduced. Because the bulk of the polymer
was nonconducting, the charge stayed where it was produced, and a stable
image of the light source resulted.

The role of the semiconductor was to assist the transfer of electrons
from the methanol to the polymer. At the end, it was unchanged, so it could
be used repeatedly.

The number of reactions which takes place depends on the amount of light
that is absorbed. The image shows this as differences in shade between
the two extremes of colour. This gives a much more realistic picture than
if the colours were sharply contrasting.

Yoneyama found that if the film was supported on an underlying electrode
the whole picture may be erased by making the electrode positive, and oxidising
the whole film so it all became yellow again.

Similar polymers could be used in renewable photocopy systems, optical
communications systems or in the etching of chips that use conducting polymers
rather than wires or semiconductors.

There are many problems to overcome, however. The polymer and its changes
in colour must be made stable for longer periods. Also, it will be necessary
to develop more coloured polymers (not everyone wants green) and to make
them stable in air. Air will oxidise the polymer and so erase the picture!