The history of automated liquid handling

Portions of this article have been reprinted with permission from Martin, J.A., Biomed Net Magazine (formerly HMS Beagle) The Art of the Pipette, April 13, 2001Copyright Elsevier.

The roots of the automated laboratory liquid handling can be traced back to the increasing focus on analytical laboratory techniques in the late 18th century. Descroizilles, a French chemist and pharmacist, invented and first introduced the buret and pipet to science in 1795. He called his burette a "berrholli-metere": it was a graduated cylinder. He later described an "alcalimetre," to dispense small volumes of liquids. Not until 1824 did Gay-Lussac publish the names "burette" and "pipette" (French, "wine tester"; Ferenc Szabadvdry, Instruments). A major breakthrough in the methodology and popularization of volumetric analysis was achieved by Karl Friedrich Mohr, who redesigned the burette by placing a clamp and a tip at the bottom. The syringe, on the other hand, was not graduated in its early stages. Early syringes were often made of metal rather than glass to avoid breakage when transported in saddlebags. A patent by Martin Overlach, "resident at Frankfort-on-the-Main, Germany" (U.S. Patent 404105, 1889), recognized a need for a syringe to house a chamber that didn't leak. Because of their mechanical nature, the syringe provided a "modernizing 'professionalization' tool," wielded by 19th-century doctors.


Gay-Lussac burette	Mohr burette	Overlach syringe	Wilson oiler

There is an ongoing technological relationship between the syringe and the eventual mechanical and automated pipets. Overlach's syringe contained pistons, and the piston remained a fundamental component of portable dispensing devices referenced in patents back to Overlach's late-19th century syringe. The syringe did not play a direct role in the development of the pipet, but the piston action, instrumental to the operation of many types of syringes, did find early success in shaping the modern pipet and eventually automated liquid handling devices.

A rash of European and U.S. patents on bicycle ("velocipede") gears appeared between 1899 and 1906, for instance, Roxendorff's "Automatic Variable Driving Gear for Velocipedes" (U.S. patent 636184), and the "Sturmey-Archer" gear. In 1903 George Wilson was granted a patent for a bicycle oiler (U.S. Patent 730065, 1903), "especially adapted for use in connection with the parts of bicycles". Wilson's oiler featured a self-closing valve that opened by way of pushing on a piston that moved down a cylinder. Wilson provided "a means for automatically returning the piston when operated to a point at or near the bottom of the reservoir."

In the late 1940's analytical instrumentation was being developed that for the first time required accurate dispensing of microliter quantities of sample. The first paper on the function and attributes of gas chromatography was published in 1944 by Professor A. J. P. Martin FRS, who was awarded the Nobel Prize in 1952 for his work in developing liquid-liquid (1941) and paper (1944) chromatography. He joined A.T. James at the National Institute of Medical Research at Mill Hill, London, in 1950 where they worked on the design and development of the first gas chromatograph and demonstrated its effective functionality in 1952.^[1]. However, there was no convenient, commercially available method to inject microliter amounts of sample into such instruments. Clark Hamilton had first developed a microliter syringe in 1947 while working at the Radiation Laboratory at U.C. Berkeley under Dr.Ernest O. Lawrence (Nobel Prize in Physics, 1939). In 1955 he decided to form The Hamilton Company to commercially produce small volume syringes^[2], which immediately became successful in the blossoming gas chromatograph market.

Hamilton microliter syringe

Mechanization

1950's - 1970's

Inventors began to find ways to automate the movement of pistons in mechanical pipets. In 1950, G.S. Riggs cited the Wilson patent in his application for a mechanical pipet (U.S. Patent 2530909, 1950). At the time, Riggs was a milk inspector. In his patent document, he noted that "it is a matter of common knowledge in the trade" that milk inspectors mouth-pipetted raw milk samples in a .018-inch-diameter glass tube and transferred the sample to a slide for view under a microscope. The Riggs' invention would have a barrel with a "suitably bored intake tip." The barrel contained a piston, "provided with a spring returned manually depressed plunger which is such as to expedite the steps of first sucking the milk." Riggs also cites the patent by Martin Overlach. J.J. Rodriguez of Berkeley, California, introduced a patent in 1961 (U.S. Patent 3013435, 1961) for an "improved buret." Rodriguez designed the buret to dispense liquids like a buret, but he claimed his invention was providing "a novel and improved buret capable of use in pipetting."

In the late 1950's, at the Institute of Physiological Chemistry at the University of Marburg, Germany, a young postdoctoral student, Heinrich Schnitger, found himself very annoyed by repeatedly using a cumbersome Carlsberg pipette (which was made by heating a glass tube over a Bunsen burner and tugging at one end to create a capillary) for transferring formic acid^[3]. Schnitger 'rebuilt' a tuberculin syringe by adding a spring to the piston that met an upward stop to define the pipetting volume. The syringe needle was replaced by a polyethylene tip, pulled from polyethylene tubing. An air buffer separated the fluid from the syringe piston and confined it to the plastic tip. Six months after he had built his first prototype, and conscious of its importance, Schnitger applied for a patent in Germany. His application, dated 3 May 1957, entitled "Vorrichtung zum schnellen und exakten Pipettieren kleiner Flüssigkeitsmengen" (Device for the fast and exact pipetting of small liquid volumes), was granted on 24 April 1961. It describes all the essential features of the modern pipette, such as the spring-loaded piston, the second coaxial spring to blow out residual liquid and the replaceable plastic tip as the sole container of the liquids. Other aspects, such as precautions to keep an air-tight seal around the piston and using an enlarged piston to keep the air cushion small and thus decreasing errors from ambient temperature effects, were also explained. Schnitger's patent application also outlined the mechanics for pipetting variable volumes either by discrete steps or by continuous adjustments. In anticipation of a broad market for the Marburg pipette, as it became commonly known—not only for research but also for medical applications—the medical supply company Eppendorf (Hamburg, Germany), bought the exclusive license for manufacturing and marketing the micropipette. In 1961 Eppendorf launched the first industrially manufactured piston-stroke pipette along with associated tips and the Eppendorf Tube. The pipette devices and tips were designed for a fixed volume dispense, so multiple models were available for different volumes.

The modern mechanically adjustable hand pipette is an invention developed through interactions among several people, including inventor Warren Gilson and Henry Lardy, a professor of biochemistry at the University of Wisconsin-Madison. Gilson Medical Electronics introduced in 1947 a modified Warburg respirometer, a device used to measure respiratory volumes. Crafted in "circular" format with a digital readout, it was a marked improvement over the original "rectangular" format Warburg respirator introduced at the turn of the twentieth century. Developed by Lardy, this device included a tiny piston used to gauge changes in the amount of oxygen. The gauge worked by using a screw to move the piston to keep the air pressure constant in a pipe as oxygen was used up. Three key things were the small size of the piston, the accuracy of the measurement and the adjustability of the volume. A piston moving into a small pipe pushed air out, and moving the piston in the other direction would suck water up. This formed the basis to create the handheld adjustable micropipette. In 1974, Gilson patented the mechanical basis for the popular adjustable pipet (U.S. Patent 3827305, 1974). He argued in his 1974 patent that Rodriguez's buret/pipet volume adjustment mechanism was too awkward and complex to adjust. Gilson also cited the mechanical features of another ancestral liquid-handling tool, G.S. Riggs 1950 mechanical pipet.


Marburg and Eppendorf pipettes	Gilson Pipetteman

Automation

1970's

By 1970's, the development of micro-scale dc motor and valve technology led to the introduction of highly accurate semi-automated motorized syringe-based pipetting devices. The Hamilton Company (described above) introduced the Digital Dilutor in 1971, a semi-automated, dc motor motor driven, adjustable pipetting device, which used two of their calibrated syringes as the pipetting pistons. The development of microprocessor technology later that decade made it possible to program sequences of motor and valve functions and this led to the development of fully automated motorized syringe-based liquid aspirating and dispensing devices. The Hamilton MicroLab M, which featured microprocessor control and a small digital display on a corded control pad, was designed to be used via a human holding the pipetting unit (connected via tubing to the bench-located syringe pump) and executing commands/methods via the keypad. Similar devices at the time include the MicroMedic Systems Digiflex.


Hamilton Digital Dilutor	Hamilton MicroLab M	MicroMedic Systems Digiflex

1980's

In this decade, the evolution of motor and microprocessor technology enabled the development of the first, true Automated Liquid Handling Workstations. The development of many of these devices were driven by the needs of the clinical radio-immunoassay market, not drug discovery. The Hamilton Company and Tecan AG had collaborated in the late 1970's on developing the Hamilton AMICA system, a microprocessor controlled stopped-flow fluidics system designed for industrial analysis, based around Hamilton motor-controlled syringes. The AMICA work and technology transfer led to the development of the Hamilton 2000 series and the Tecan Sampler 500 / RSP 5000 series workstations^[4]^[5]. Both were based on a cartesian robotic platform, featuring a single washable pipetting channel attached to the cartesian arm, connected via inert tubing to a microprocessor-controlled, dc motor driven syringe pump. Shortly thereafter, these platforms became available with a second cartesian arm and a second single pipetting channel. The Zymark Z510 Master Laboratory Station was introduced in 1982 as a 3x motor-driven syringe unit, designed specifically to be controlled remotely, as part of a larger Zymate integrated robotic system (marketed toward the analytical laboratory / HPLC sample prep market)^[3]. The Zymark Zymate laboratory robot system also offered a "syringe hand", a robotic end effector with an internal motor to drive a commercially available glass syringe, which enabled the robotic arm to execute single channel liquid transfers with needles or pipette tips. From 1982 to 1984 a very small company, Infinitek, had been developing a robotic liquid handling workstation that was specifically aimed at drug discovery microplate assay development ^[6]. Based on a "breadboard" prototype, the company identified 13 initial customers, with two (Genentech and Chiron) willing to pay a deposit. Faced with a daunting remaining amount of development (estimated at two years) but with solid market interest, the Infinitek owners sold the company and workstation design to Beckman Coulter (then Smithkline Beckman). The ensuing Biomek 1000, launched a year later, was based on a combination of cartesian robotic arm and positioning stage base. The Biomek featured interchangable multi-channel, pipette tip-based liquid handling tools. Also, in 1985, Kemble (U.K.) marketed a cartesian liquid handling workstation, the Star 700. Two years later, Packard Instrument Company would market the Probe, an OEM cartesian workstation from Kemble.

The Cetus Propette appeared in 1986 as a small, 12-channel, multi-microtiter plate liquid transfer device focused solely on pipette tip based microplate transfers and cannula-based plate washing. It was originally designed by Cetus engineeers in response to the needs of Cetus scientists for automating Interleukin-2 assays, but also became very popular for doing PCR work^[7]. It consisted of a single axis positioning stage for holding microplates, pipette tips and reservoirs, and pipetting and washing heads mounted above on a linear z-axis. Subsequently, it was marketed to the rest of the biotech community via a joint venture with Perkin Elmer. In 1987 Hamilton introduced the Microlab AT, desgined to do batch screening of whole blood for aids and hepatitis viruses^[4]. The AT featured variable span, 12 channel, disposable tip pipetting. In 1990, TomTec Inc. developed the first 96-channel pipetting/liquid transfer device, the Quadra96, which had a positioning stage base for holding microplates, pipette tips and reservoirs, with a 96-channel pipetting head mounted above on a linar z-axis.^[8].


Hamilton MicroLab 2200	Tecan Sampler 500	Zymark Z510 Master Lab Station	Zymate Syringe Hand	Beckman Biomek 1000


Cetus Propette	Hamilton MicroLab AT	TomTec Quadra96

The current Automated Liquid Handling Workstations have all evolved from these early basic designs, adding LUO functionality and software sophistication.

External links

References

↑ Nobel Lecture of Archer J.P. Martin, 1952
↑ Clark Hamilton is 70 years old, Chromatographia, 15, 333, 1982
↑ ^3.0 ^3.1 | Martin Klingenberg, When a common problem meets an ingenious mind, European Molecular Biology Organization reports
↑ ^4.0 ^4.1 Personal communication, Steven T. Hamilton and Jim Salika, The Hamilton Company.
↑ Tecan History, Tecan Group Ltd.
↑ http://www.productdevelopment.com/customers/cs-beckmancoulter.html
↑ Biotechnology: the science and the business, Vivian Moses, D. G. Springham, Ronald E. Cape, CRC Press, 1999
↑ Small volume pipetting