FIELD OF THE INVENTION

The present invention relates to process and device for minimally invasively tissue probing, including sampling and/or injecting. The invention further relates to a process for shaping materials into hollow or solid microneedles for the sampling and/or injection, and for their use. The present invention further relates to the use of the solid devices for light guidance.

BACKGROUND OF THE INVENTION

In recent years, many operations in sensitive tissues and organs, such as for instance deep brain tissue are performed through stereotactic endoscopic surgery, using probing and injection instruments of ever-decreasing size. Needle assemblies are commonly used to either inject substances into or extract substances out of human or animal tissue.

However, most needles presently employed are made from drawn stainless steel needles, such as for instance those disclosed in US-A-2, 814,296. A needle cannula is typically drawn from stainless steel through progressively smaller dies to reduce the diameter at the needle tip. The ratio between the outside diameter and the inside diameter is then a result of the wall thickness of the stainless tube drawn and the diameter to which it is drawn, but generally, is diminished over the length of the drawing process. Needle tips are then usually prepared by grinding or similar mechanical operations. The smallest commercially available needles, so called 34G thin-wall needles, typically have an outer diameter (OD) of 200 pm, and an inner diameter (ID) of 100 pm, but have comparatively blunt tips.

The commercially available needles with small tip diameters are generally prepared by heating a middle section of a glass cannula while pulling both ends of the cannula outwards. The thus drawn needles may than be cut, or may be etched to separate and the two needles.

While this can lead to sub micrometer tip diameter, the inner and outer diameter of the cannulae usually rapidly decreases in the direction of the tip, resulting in a high pressure drop over the needle tip in use. The funnel shaped lumen is also prone to clogging. The volume of the lumen thus created is also very high compared to the tip dimensions. US-A-2007282265 discloses hollow needles made from ceramic, specifically zirconium oxide, having the outer diameter of 1 to 5 mm. The needles have a wall thickness of from 0.3 mm to 1 .8 mm, and are prepared by either covering a wire with ceramic powder, and then burning of the powder into a ceramic, or by mould injection and burning. While this process allows preparation of comparatively tough needles that can be employed where equally dimensioned steel needles would buckle under the pressure during injection into tougher tissues, the thus obtained needles still have the same, large dimensions, resulting in tissue damage. Yet further, the process is tedious and complex.

The use of the above disclosed needles may lead to physical damage of the tissue during the entry and exit of a needle, and to haemorrhage, and thus to the destruction of vital cells in the path of the needle. This is in particular relevant for brain or nerve tissue probing, where even the smallest damage can be detrimental for nerve cell activity in the impacted areas. It may, depending on the damage exerted, even require opening of the area of the incision completely to repair damaged nerves or vessels.

Further issues with existing needles include that due to the changing diameter of the lumen of the needle due to the drawing process, or due to imperfections on the surface of a ceramic needle, pressure may build up during injections, clogging of the cannula may occur, and a less favourable shape for sampling is obtained. Yet further, as conventional small needles as described herein above are sharpened by bevelling the tip, this limits the depth accuracy when probing.

Accordingly, there is a need for deep tissue probes, whose use results in significantly less trauma, as well as a suitable versatile process for their manufacture.

SUMMARY OF THE INVENTION

Accordingly, in a first aspect, the present invention relates to an ultra- high aspect ratio needle cannula with pointed, distal end having an inner diameter in the range of from 0.5 to 250 pm, preferably 5 to 50 pm, an outer diameter in the range of from 50 to 2500 pm, and an aspect ratio of tip to length in the range of from 100 to 1x10 ¹⁰ , for use in applications including deep tissue probing, painless injections, fluid dispensing, electrospray ionisation and light guiding.

In a second aspect, the present invention also relates to a process for the manufacture of a high aspect ratio needle cannula with pointed distal end deep tissue probing, fluid dispensing, electrospray ionisation and/or light guiding, comprising the steps of: (i) providing a needle cannula having a proximal end and a distal end and an oblong inside lumen there between;

(ii) at least partly covering the outer surface of the cannula with a protective coating which is inert to the etching liquid, such that the coating prevents the etchant from directly contacting the outer surface of the cannula;

(iii) at least partly preventing the etchant from entering the lumen or etching the inner surface, preferably by coating an inside surface of the lumen with a protective material, which is inert to the etching liquid, or by preventing the etchant liquid from entering the lumen, such as by applying a gas flow with sufficiently high pressure to the lumen, and (iv) contacting the front end of the cannula with an etching liquid to obtain a needle cannula with a pointed distal end. Preferably the process further comprises a step (v) at least partly coating an outside surface of the needle cannula with a coating material, prior to providing the etching liquid. The outside coating material may be chosen from any suitable material preferably a polymeric material; which improve mechanical properties of the cannula, such as polyimides. The coating material may preferably be applied in a range of 5 pm to 500 pm thickness.

The coating preferably is inert to the etchant, however permits the etchant to slowly diffuse through the coating. Without wishing to be bound to any particular theory, it is believed that the partial permeability of the coating may allow the needles to be etched through the coating, after which the etchant is sucked up between de fused silica and the coating by capillary force, thereby resulting in a particularly high aspect ratio. Without the coating the needle will primarily be etched where submerged in the etchant.

Description of the Figures

Figure 1 : Figure 1 depicts light microscopy images of a sharpened needle cannulae resulting from the process according to the invention.

Figure 2: Figure 2 depicts micrographs of the formation of a needle cannula over time.

Figure 3: Figure 3 depicts micrographs of rat brain tissue with injections according to the process and with a needle according to the invention, as well as comparative injection.

Figure 4: Figure 4 discloses a batch of 81 fused silica capillaries mounted in a polyethylene holder and placed over a Teflon container with 49% HF. All needles are etched simultaneously by immersion in 49% HF. Immersion depth can be adjusted by the HF level in the container and the screw positions of the holder during mounting Figure 5: Figure 5 depicts a microneedle applicator was developed using an electromagnet (a) to insert a microneedle (c) into the skin at controlled speed. A micrometer actuator (b) is used to adjust the level of a supporting plateau (d) to ensure proper positioning and penetration depth of the microneedle. A graduated rail (e) can be used to tilt the entire applicator for angled insertion. The microneedle is connected to a syringe pump with a capillary (f) for vaccine delivery, g) Close-up of the supporting plateau with a through hole for the needle, h) Microneedle applicator injecting a vaccine into the skin of an anesthetized rat.

Figure 6: Figure 6 discloses the determination of the depth of microneedle insertion. After hollow microneedles were pierced into fluorescein labeled human skin, one picture of the hollow microneedle was taken by using bright field microscopy (A), and another picture on the same position was taken by fluorescence microscopy (B). Subsequently, the bright field and fluorescence microscopy pictures were overlaid (C) from which the depth of microneedle insertion was measured and plotted as a function of the aimed insertion depth (D). Each point in panel D represents the average (± SD) of three individual microneedle insertions.

Figure 7: Figure 7 discloses the fluorescence visualization of fluorescein microinjections into ex vivo human skin. Control injections (fluorescein dispensed on top of the skin) result in solvent evaporation over time and, consequently, loss of fluorescence, while 300 pm deep injections show no leakage and loss off fluorescence, but prolonged fluorescence as well as diffusion into surrounding tissue.

Figure 8: Figure 8 shows the serum IgG responses (mean±SEM) after immunization with PBS or inactivated poliovirus serotype 1 intramuscularly (i.m., 200 μΙ_), intradermal^ (i.d., 50 μΙ_), or intradermal^ via hollow microneedles with a microinjection of 9 μΙ_ containing either 5 or 15 DU IPV.

Figure 9: Figure 9 depicts serum polio-virus neutralizing (VN) antibodies responses after immunization with PBS or inactivated poliovirus serotype 1 intramuscularly (i.m.), intradermal^ (i.d.), or intradermal^ via hollow microneedles with a microinjection of 9 μΙ_ containing either 5 or 15 DU IPV.

The present invention encompasses a hollow needle shaped cannula with a distal end having preferably an atom sharp edge.

The proximal end is typically comprised of a cylindrical or otherwise shaped tube. The inner diameter of the tube may be at least 0.5, such as at least 1 , or at least 10, or at least 20, or at least 50 μηι in diameter. The diameter, and hence lumen volume may be selected according to a particular application. For fluid injection or sampling of reasonable volumes, an inner diameter of at least 15 pm is preferred.

The outer diameter is preferably at most 500 pm, more preferably at most 450 pm, yet more preferably at most 350 pm, again more preferably at most 300 pm, and yet more preferably at most 250 pm, and most preferably at most 200 pm,

Depending on the process, the outer surface of the cannula may taper over a length of hundreds of micrometres up to tens of millimetres to an atom sharp edged tip at the distal end. Advantageously, a generally constant inner diameter lumen may run along the length of the cannula from the proximal to the distal end allowing for the transport of e.g. liquids or gasses through the needle, avoiding for instance an undesired pressure drop.

The length and angle of the tapered part of the needle can preferably be applied to a wide range, ranging from several pm lengths to several centimetres.

Preferably the outer diameter of the proximal end can be chosen between 10 pm and several millimetres. The internal diameter can range from 0.1 pm to several millimetres. Even at relatively high inner and outer diameter, the subject invention provides for a very sharp edge at the tip, thereby diminishing negative effects during the deep tissue probing.

The high aspect ratio of the needle and the atom sharp tip advantageously allow for easy penetration into a substrate with minimal damage to the substrate, for less invasive, local sampling or injection. Again further, the comparatively extremely high angle of the edge at the tip will prevent wetting of the sides of the needle when it is used as a nanoelectrospray ionisation source, and hence the subject invention also relates to this use. The extreme angle of the tip, as expressed by the high aspect ratio, will also help preventing wetting of the outside of the canulla when fluids are dispensed from the cannula.

Furthermore, the small diameter of the tip allow for the use of the needles with exceedingly small substrates such as single cells or specific skin or tissue regions.

The lack of a high angle bevelled tip also increases the depth accurately allowing for accurate injection to a certain depth, while also preventing leakage in shallow injections. Yet further, the lumen preferably has a constant internal diameter preventing clogging that is associated with conventional needles with a tapered lumen. The constant lumen further also advantageously improves capillary filling rates when sampling as the small diameter ensure high capillary suction along the entire length of the lumen.

US-A-4, 469,554 discloses a process comprising at least partially immersing a cylindncally symmetric body in first liquid in order to etch the cylindrically symmetric body characterized in that a second liquid is located on top of the first liquid, said second liquid being substantially non-etching and substantially immiscible in the first liquid and the cylindrically symmetric body is a glass optical fiber consisting of at least 80 weight percent silicon dioxide.

The publication entitled "Fabrication Process of Microsurgical Tools for Single- Cell Trapping and Intracytoplasmic Injection", JMEMS, p.940-945, Vol.13, No. 6, Dec. 2004, discloses the preparation of single-cell trappers and microinjectors from fused silica capillaries via a diffusion limited etching method based on US-A-4, 469, 554, and their use as tools for single cell manipulations. However, in this process, the entire length of the tip is immersed in the etching liquid, thereby limiting the available aspect ratio.

Preferably the material of the cannula is essentially an isotropic, etchable composition, preferably fused silica, glass and/or pure metals.

Generally, the material subjected to etching should be an isotropic material with appropriate physical and chemical properties. Suitable examples are glass, and pure metals such titanium and/or copper. The glasses may have different physico-chemical properties, including of varying density, strength, optical transparency at a certain wavelength.

Other suitable materials include ceramics, such as zirconia, titania and alumina. The ceramics may be conductive. Further suitable materials include polymeric materials, which may be etched with a suitable oxidative etchant or solvent.

Alloys or similar component comprising different domains have proven less successful since e.g. in steel, specific crystalline domains are etched away faster than other, leading to a pitted surface and a structurally weaker composition. Applicants found that for instance fused silica, preferably with an polyimide coating, proved sufficiently strong to not break upon deep tissue insertion, nor in the retraction of the needle through repeated use.

The substrates may be round, elliptical, rectangular, square, triangular, or of any other suitable shape.

In the process according to the subject invention, various etchants may be used in the practice of the invention. Generally, the etchant should etch, or dissolve the capillary material being etched at a reasonable rate and approximately isotropically, although rotation of the material around the cylindrical axis might preferably increase the isotropic nature of the process.

Typical etchants include aqueous HF for titanium, fused silica and various glasses including quartz glass; and aqueous sodium perchlorate for copper, or aqua regia for gold and platinum.

Step (iii) is preferably performed by immersion of the distal end of the cannula in the etchant, i.e. the capillary tubing is preferably immersed into an etchant bath.

The protective material used in the inert cavity should preferably not be miscible with the etching solution or etchant, or only to a very small extent to avoid the interior of the capillary to be in contact excessively with the etchant.

The protective material should further be essentially chemically inert to the etchant. A limited amount of chemical attack may be tolerable, but materials that are prone to excessive attack should be avoided. Yet further, the protective material should preferably have a low vapour pressure to avoid evaporation of the liquid.

Typical protective materials that are useful in the present process include silicon oils, hydrocarbon oils and waxes; and/or polyolefins. The protective materials preferably are liquid or can be liquefied such that they can be easily introduced into the capillary tubing prior to the etching process.

Upon immersion, the etchant wets the surface of the capillary that is non-coated.

Where the capillary is provided with a coating, inside the coating, the etching process will be determined by the mass transport of the etchant up the capillary.

Without wishing to be bound to any particular theory, the lowest part of the immersed capillary is in close contact with unsaturated etchant, resulting in a faster etch. Further up the capillary, the etchant needs to diffuse further along the coating to reach the capillary material, resulting in a lower concentration of etchant, and hence slower etching. As a result, a particularly sharp edge is obtained at the tip of the capillary, leading to an atom sharp tip. Furthermore, etching through the coating appears to cause thinning of the entire length of the capillary as immersed or exposed to vapour. The protective material therefore preferably permits etchant vapours to diffuse through the material to the surface of the canulla. The combination of the direct contact at the tip, and the exposure to a lower concentration of the etchant in the non-immersed part through the protective material was found to contribute to the efficacy, and resulted in particularly high aspect ratios attainable by this method. The cannula according to the invention may advantageously be produced by etching sections of the capillary tubing with an etchant. Before etching, the inner surface of the lumen of the tubing is preferably coated, or completely filled with a liquid protective material such as for instance silicone oil, liquefied wax or a hydrocarbon oil. This may conveniently be done by immersing the same end into the solvent and allowing the capillary forces to draw the oil inside, preferably assisted by applying a vacuum to the proximal end if the amount and/or penetration or coating depth of protective material is more than that achieved through capillary forces. This step assures that the inner surface of the capillaries where in contact with the etchant is protected from the etchant during the subsequent etching processes, thus preserving the inner geometry. The protected capillary is then at least partially immersed into an etchant solution. If the etchant is not allowed to enter into the capillary, e.g. by a plug of oil, it is preferably not necessarily needed to coat the entire lumen.

The outside of the capillary may advantageously be covered with a protective coating. This coating, while not necessarily being inert under the conditions of the etching process, may serve as a barrier to the etchant to reduce the rate of etching on the outer surface of the tubing, thereby resulting in a higher cannula aspect.

The outside coating is in particular relevant where a high aspect ratio cannula is required, where the tip is longer than the height of the meniscus that would be achieved in the case of a non-coated needle. Without a protective coating, only the meniscus will determine the shape and length of the tip, since the entire submersed capillary is etched away.

Suitable coatings include thermoset flexible coatings, preferably a polyimide, a fuoropolymer and/or a dimethylsiloxane coating. Preferably the exterior coating is a flexible, protective coating that encloses the exterior of the capillary for substantially the entire length of the capillary.

The protective coating is preferably applied to the capillary during manufacturing of the capillary. Typical coatings include the polyimide coatings applied for instance to gas chromatographic columns.

Without wishing to be bound to any particular theory, it is believed that the surface tension causes the etchant to partially displace the protective fluid, e.g. silicon oil, from inside the capillary. In order to ensure the fabrication of the sharp tip, the capillary is immersed to a depth such that the level of the etchant/protective material interface inside the capillary is lower than the contact line between the etchant and capillary on the outside.

While the protective coating ensures that the walls above the actual tip stay intact, it should preferably allow a slow etching process that enables thinning of the walls along a certain length of the needles, thus decreasing the outer diameter, and increasing the aspect ratio.

As a result, the etchant will remove a very small length of the capillary wall material from the inside out, as well as a small amount from the outside. The remaining portion of the capillary may be advantageously filled with protective material, such as inert material and/or a protective gas, which preserves the inner geometry during the etching process.

The etching process is performed at suitable conditions. In the case of fused silica, this may conveniently be performed at room temperature. The etching rate can be estimated from measuring the end diameters of the etched section of the capillaries at different times. Accurate determination of the etching rate is useful to achieve high repeatability.

Preferably, the aspect ratio of cannula outer diameter and length is in the range of from preferably 1 : 1000, more preferably 1 : 100.000, yet more preferably 1x10 ⁶ , yet more preferably 1x10 ⁷ , yet more preferably 1x10 ⁸ , again more preferably 1x10 ⁹ , and most preferably 1x10 ¹⁰ .

The term "aspect ratio" herein denotes the aspect ratio of the width \N as measured at the tip of the pointed distal end to the length L measured from the tip to the beginning of a continuous contraction of the outer diameter of the cannula width; i.e. the length from the distal end of the tip to the proximal end of the tip i.e. the basis where the outer diameter starts to decrease. The thickness of the tip in the case of solid needle, or the wall thickness in the case of a hollow needle may be measured by any suitable method; where ultra high aspects are achieved, this may require the use of SEM or tapping atomic force microscopy.

Preferably the ratio is in the range of from 1 10 to 900.000, Yet more preferably, the ratio is at least at or above 120, 130, 150, 180, 200, 300, 400, 500, 800, 1000, 1200. Yet more preferably, the ratio is at most at or below 850.000, 800.000, 750.000, 700.000, 500.000, 450.000, 300.000, 250.000, 150.000, 100.000, 500.000.

The cannula according to the present invention furthermore may have a particularly sharp tip, i.e. a tip having a width between outer and inner wall of the cannula in the range of from one to several atom radii. Accordingly, the distal tip of the wall may have a wall thickness at the tip of from 1 10 to 1500 pm, more preferably of from 150 to 500 pm. This advantageously allows penetrating materials, such as tissue, with minimal damage, and/or without inflicting pain.

Preferably, the inner diameter of the cannula is preferably in the range of from 5 to 50 pm. Furthermore, the outer diameter of the cannula is preferably in the range of from 10 to 2500 pm, and preferably tapering to a very sharp tip, as set out below

The subject invention further relates to a high aspect ratio needle cannula with pointed, tissue penetrating distal end for deep tissue probing, having an inner diameter in the range of from 5 to 50 pm, an outer diameter in the range of from 50 to 2500 pm.

The thus obtained cannula may preferably be coated with conductive layers, releasable coatings, affinity coatings, e.g. antibodies, coatings that change the refractive index of the outer layer to keep light inside an optical conduit. Suitable coatings are those that make a needle visible with MRI or X-ray during surgery to allow very accurate positioning. Furthermore, overplating of the needle by a metal such as Nickel may not only allow to form a conductive coating, but at the same time may increase stability of the needle and/or the coating. Yet further, coatings may be applied that strengthen the capillary, increasing its flexibility under pressure, and/or reducing the tendency of breaking off. The invention further relates to a device for deep tissue probing, comprising the needle according to the invention.

Furthermore, the surface of the outside or inside of the needle wall can be modified. This may advantageously allow for tuning the rigidity of the needle, to provide, or enhance electrical conductivity allowing the needle to be used as an electrode, or to attract or release certain compounds.

In a further aspect, the invention further relates to a process for probing or injection into deep tissue, comprising a) providing a needle according to the invention, and b) inserting the needle into a tissue to inject material and/or extract material. The insertion may also be performed to selectively measure the release of certain metabolites, and/or to diagnose certain conditions, which may also be done in vivo, as well as for the stimulation of certain tissues.

The present process further relates to a method of where the needle is coated with a conductive layer so an electrical response, e.g. in brain tissue can be measured directly, or after a certain compound has been injected with the needle into the brain tissue. The invention further relates to a multitude of cannulae assembled into a microarray of high aspect needles, and to devices comprising one or more cannulae according to the invention.

The present invention further advantageously relates to a process for handling tissue material, comprising a) providing one or more cannulae according to the invention, and b) providing tissue material; and c) handling the tissue material using the one or more cannulae. Preferably, the handling of tissue material comprises probing and/or injection of an injectable material into the tissue material.

The present invention further advantageously relates to a process for recording and/or stimulating intracellular activity, comprising d) filling the cannulae with a conducting medium and connecting the cannula to a measurement apparatus; or using a conductive cannula connected to a measurement apparatus; e) penetrating a cell membrane, and f) recording and/or stimulating cell activity.

The present invention also relates to a tool for supporting one or more of cannulae when subjected to the process according to the invention, comprising one or more flexible parts , each of which seals an area of the outside surface of each cannula from being in contact with the etching liquid or vapours during the execution of the etching step.

The present invention also relates to the use of one or more hollow cannulae according to the invention, or obtainable according to the process according to invention as tissue probing device, nanoelectrospray ionisation source, painless injection needle, for the handling of tissue material, recording or stimulation of tissue or cells, for the ocular or single cell probing, including IVF, and/or for release of compounds into tissue.

The present invention also relates to the use of one or more solid cannulae according to the invention, for light transfer and/or focussing, and/or light based stimulation or recording.

The finished needle may further preferably be subjected to a secondary etch to make the material porous, resulting in cannulae useful as electrospray emitters.

Yet further, the present invention also relates to the use of a cannula for deposition, for use of the needle for removal of compounds (after injection, after dipping into a medium and from a fluid like air or gas); and as a non-clogging ESI emitter.

In a further preferred embodiment, applying a coating over the needle may allow for the formation of a membrane at the distal end of the tip, permitting the use of such a cannula for microdialysis and electro-microdialysis. Conventional intramuscular or subcutaneous injections of vaccines cause pain and stress, carry the risk of infection and need trained personnel. By only piercing the top layers of the skin, microneedles can be used for minimally invasive, pain free vaccine delivery to an immunologically active site.

The subject invention also relates to a method for for painless intradermal vaccine delivery using the hollow microneedles according to the invention. In this methid, one or more microneedles are preferably used in combination with an automated applicator, for instance advantageously electromagnetic applicator to control the insertion speed, depth, and or angle for painless intradermal vaccine delivery.

The following, non-limiting examples illustrate the invention:

Example 1 : Small Inner Diameter Needle

A needle with a cylindrical inner geometry was fabricated by chemical etching in a diffusion-rate limited regime. The fabrication employed a polyimide coated, fused silica tubing usually employed for gas chromatography. First, the capillary was filled with silicon oil by capillary force. The capillary was then immersed into 49%wt. HF solution in a depth of several centimetres, for a period of several hours at room temperature, while capillary above the etchant was exposed to the gaseous atmosphere above the etchant.

After 2 hours of immersion, the capillary delaminates from polyimider coating as the spce between the fused silica and the polyimide becomes larger. During the next hours, this intermediate space increases. In the periode off from 1 1 to 18 uur etching ultra high aspect ratio needles are formed (Figure 2).

Example 2: Intracerebral Injection into Rat Brains

A needle according to example 1 was connected to a injection device, and a coloured fluid was injected into rat brain tissue. As comparative example, a commercially available and widely applied injection needle was employed.

The results are depicted in Fig. 3, showing a cut of brain tissue after three different injections into ventricles, and injecting Evans Blue to show the damage through the injection (track 2 as the result of a conventional small needle, tracks 1 and 3 by needles according to the subject invention).

Example 3: Subcutaneous injections of vaccines

Hollow microneedles with an inner diameter of 20 pm were used to immunize rats with an inactivated poliovirus vaccine by an intradermal microinjection of 9 μΙ_ at a depth of 300 pm and an insertion speed of 1 m/s. Intradermal microinjection induced comparable immune responses to conventional intramuscular injection, demonstrating the potential of the subject microneedles for pain free, minimally invasive intradermal vaccination.

Polyimide coated fused silica capillaries (Polymicro, Phoenix AZ, USA, 375 pm outer diameter, 20 pm inner diameter) were wet etched into microneedles. First, the lumen of the capillaries was filled with silicone oil AK350 (Boom Chemicals, Meppel, the Netherlands) to protect the inside of the capillary from the etchant. Subsequently, to etch the capillaries into microneedles, batches of up to 81 capillaries were mounted in a holder with one end suspended in 49% (w/w) HF, as shown in figure 4. The length of the capillary that is immersed in HF could be adjusted by the level of HF in the container as well as by the position of the screws of the holder during mounting. The shape of the microneedle tip was investigated as a function of etching time up to 30 hours. The microneedles chosen for the intradermal injections were etched for 4 hours. In order to connect the microneedles to the applicator, the sharp capillaries were glued into Luer adapters using medical grade glue (Loctite M-31 CL, Henkel, Dusseldorf, Germany). Luer adapters were acquired by removing the needle from Microlance® 26G hypodermic needles (BD, Franklin Lakes, NJ, USA). Finally, the polyimide coating was removed from the microneedles by submersion of the etched capillaries into hot sulfuric acid (96-98%, Boom chemicals, Meppel, the Netherlands) for 5 min at 200°C, and the silicone oil was removed by flushing the microneedles with acetone.

In order to enable insertion of the microneedles into the skin to a controlled depth in a reproducible way, an applicator for hollow microneedles was developed. The flexibility of the skin can hamper skin penetration and reduce the depth accuracy of injections. The insertion speed, insertion depth, and angle of insertion are important factors to overcome this and achieve successful depth controlled penetration and drug delivery. The applicator shown in Figure 5 used an electromagnet (a) to insert the microneedle mounted on the Luer connection (c) into the skin. The electromagnet was actuated using a controllable power source. Application of a variable current through the electromagnet allows for a tuned insertion speed from 1 -3 m/s. Furthermore, the controlled power supply regulates the wear time, to accurately and reproducibly control the time the microneedle is kept inside the skin. For injection, the skin is placed on a platform positioned below the injector (g). The penetration depth is controlled by a supporting plateau (d) that can be adjusted with a micrometer actuator (b). The position of the supporting plateau is adjusted so that the microneedle protrudes the desired distance through the hole in the plateau when the electromagnet is actuated. Before injection the injector is lowered so the supporting plateau rests on top of the skin. Upon actuation the needle is inserted into the skin at the desired speed and depth, as controlled by the current and plateau position. Additionally the entire applicator can be tilted along a graduated rail (e) for angled insertion (15°-90° relative to the skin). Insertion at a shallow angle can be used to insert a greater length of the needle into the skin, without penetration into deeper skin layers. The hollow microneedles were attached to a syringe pump that was equipped with a gas tight Luer lock 100 μΙ_ syringe and connected to a fused silica capillary with a diameter of 100 pm (f) to allow tuning of the liquid delivery rate.

To determine the microneedle insertion depth, fluorescently labeled human skin was used. Ex vitro abdominal or mammary human skin, which was obtained within 24 h after cosmetic surgery, was dermatomed to a thickness of 1200 pm using a Padgett Electro Dermatome Model B (Kansas City, MO, USA) after the fat was removed. The dermatomed skin was incubated overnight on filter paper soaked with 10 pg/mL fluorescein (Fluka®, Sigma-Aldrich, Steinheim, Germany) with the stratum corneum side to the surface in a Petri dish at 4°C. Before the skin was used for experiments, it was stretched on a piece of expanded polystyrene covered with parafilm, and thereafter the surface of the skin was cleaned 3 times with phosphate buffered saline (PBS) (Braun, Melsungen, Germany, pH 7.4, 163.9 mM Na+, 140.3 mM CI-, 8.7 mM HPO42-, and H2PO4-). Subsequently, a microneedle was connected to the Luer connection on the microneedle applicator (figure 5c) and the micrometer actuator (figure 5d) was adjusted to a chosen depth up to 400 pm. In order to accurately set the insertion depth, the microneedle position was calibrated. First, the electromagnet was actuated and the microneedle was positioned to be flush with the plateau using the micrometer actuator (figure 5a). This was done by laser alignment, by which a laser was put perpendicular to the supporting plateau of the applicator (figure 5d). When a microneedle protrudes past the supporting plateau the laser light is scattered at the tip of the microneedle, therefore the needle was determined to be flushed with the plateau at the point where no light scattering was observed at the microneedle tip. After this calibration of the microneedle position, the microneedle actuator was adjusted to a depth of 100, 200, 300, or 400 pm, and was subsequently pierced into the fluorescently labeled human skin. For each chosen depth 3 microneedles were used to pierce the skin and each microneedle stayed in the skin for 5 s. After the microneedles had pierced the fluorescently labeled skin, the microneedles were analyzed by bright field microscopy and by fluorescence microscopy (Nikon Eclipse E600, mercury light source, GFP filter set) at 100x magnification and imaged with a 1 s exposure time. Subsequently, the bright field and fluorescence microscopy images were overlaid in ImageJ (available from rsbweb.nih.gov/ij/), and the insertion depth was expressed as the measured length between the tip of the microneedle and the part up to where the fluorescence was visible.

Microinjection into skin

To visualize the intradermal injections, a 10 pg/mL fluorescein solution in PBS was injected into human ex vivo skin at different depths (100-400 pm). The skin was stretched on Styrofoam covered with parafilm, and the injection depth was adjusted as described above. The microneedle was connected to a syringe pump with a flow rate of 2 pL/min to deliver 3 μΙ_ of the fluorescein solution. As a negative control 3 μΙ_ of the fluorescein solution was dispensed on top of the skin. Furthermore, diffusion of the injected compound into surrounding tissue was assessed by imaging the fluorescence in the skin at several time points after injection. The fluorescence was imaged using a Microsoft Lifecam HD with a 480±5 nm light source (LED, Conrad) and a 520-540 nm band pass interference filter.

Immunization

To examine the applicability of the hollow microneedles for vaccination purposes, a comparison was made between vaccination of rats by intradermal injection with microneedles, and intradermal and intramuscular injection with a conventional needle.

Female Wistar Han rats (175-250 g) were obtained from Charles River (Maastricht, the Netherlands) and were maintained under standardized conditions in the animal facility of the Leiden Academic Centre for Drug Research, Leiden University. The study was carried out under the guidelines complied by the animal ethic committee of the Netherlands. Rats (10 per group) were immunized with 5 DU (1/8th human dose) IPV-type-1 (1 DU«13 ng virus protein, obtained according to Westdijk, J., et al., Characterization and standardization of Sabin based inactivated polio vaccine: Proposal for a new antigen unit for inactivated polio vaccines. Vaccine, 201 1 . 29: p. 3390-3397) intradermal^ by hollow microneedles at an apparent depth of 300 pm and a flow speed of 2 pL/min for 4.5 minutes. For comparison, rats were immunized with 5 DU IPV/50 μί intradermal^ on the belly with a 30G needle and with 5 DU IPV/200 μί intramuscularly with a 26G needle (100 L/hind leg). As a negative control rats were injected intramuscularly with 200 μί PBS (100 pL/leg). All rats were anesthetized during the immunization with 50 mg/kg ketamine (Nimatek® (100 mg/mL Ketamine, Eurovet Animal Health B.V., Bladel, the Netherlands)) and 5 mg/kg xylazine (Rompun® (20 mg/mL xylazine, Bayer B.V., Mijdrecht, the Netherlands)) by intraperitoneal injection. Three weeks after immunization blood samples were collected and the rats were sacrificed. This immunization scheme is identical to the regular IPV potency assay, as described in an Steenis, G., A. van Wezel, and V. Sekhuis, "Potency testing of killed polio vaccine in rats", Developments in biological standardization, 1981. 47: p. 1 19-128.

IgG ELISA

Polystyrene 96 wells microtiter plates (Greiner Bio-One, Alphen a/d Rijn, the Netherlands) were coated overnight at 4°C with bovine anti-poliovirus type 1 serum (RIVM, Bilthoven, The Netherlands) in PBS pH 7.2 (Gibco from Invitrogen, Paisley, UK). After washing with 0.05% Tween 80 (Merck, Darmstadt, Germany) in tap water, 100 μΙ/well monovalent IPV vaccine type 1 (4.5 DU/well) diluted in assay buffer, PBS containing 0.5% (w/v) Protifar (Nutricia, Zoetermeer, the Netherlands) and 0.05% (v/v) Tween 80 (Merck, Darmstadt, Germany), was added. After incubation at 37°C for 2 hours, threefold dilutions of sera samples in assay buffer were added (100 μΙ/well) and incubated at 37°C for 2 hours. After washing, plates were subsequently incubated at 37°C for 1 hour with horseradish peroxidase (HRPO)-conjugated goat-anti-rat IgG (Southern Biotech, Birmingham, AL) as detection antibody (4000 fold dilution, 100 μΙ/well). Plates were extensively washed and 100 μΙ/well TMB substrate solution, containing 1 .1 M sodium acetate (NVI, Bilthoven, the Netherlands), 100 mg/ml 3, 3', 5,5'- tetramethylbenzidine (Sigma-Aldrich, St. Louis, MO), and 0.006% (v/v) hydrogen peroxide (Merck, Darmstadt, Germany), was added. After 10-15 minutes, the reaction was stopped with 100 μ l/well 2M H2SO4 (NVI, Bilthoven, the Netherlands) and absorbance was measured at 450 nm by using a Biotek L808 plate reader.

Endpoint titers were determined by 4-parameter analysis using the Gen5™ 2.0 Data Analysis software (BioTek Instruments, Inc., Winooski, VT) and defined as the reciprocal of the serum dilution producing a signal identical to that of antibody-negative serum samples at the same dilution plus three times the standard deviation, as disclosed in Westdijk, J., et al. cited above, and Westdijk, J., et al., Antigen sparing with adjuvanted inactivated polio vaccine based on Sabin strains. Vaccine, 2013. 31 : p. 1298-1304.

Virus neutralizing antibodies Virus neutralizing (VN) antibodies against poliovirus type 1 were measured as described in Westdijk, J., et al., above. In brief, the sera were inactivated at 56°C for 30 min prior to testing, of which 2 fold serial dilutions were made (21 to 224). These dilutions were incubated with 100 cell culture infectious dose 50% (CCID50) of wild type poliovirus type 1 (Mahoney) for 3 hours at 36°C and 5% C02, and the resulting mixtures were subsequently used to inoculate 1 ·104 Vero cells. Virus-neutralizing antibody titers were determined after 7 days by staining the cells with crystal violet and expressed as the last serial dilution with an intact monolayer.

Statistical analysis

The statistical analysis was performed using Prism 5 for Windows, where the mean ± SEM (n=10) of the IgG and VN titers were calculated. Furthermore, statistical significance was determined by a two way analysis of variance (ANOVA) with a Bonferroni post test.

Production of microneedles: Influence of etch time

The microneedle geometry was observed to be dependent on etch time. The different tip shapes that were obtained after 1 , 2, 4, 9, 22 and 29 hours of etching are depicted in figure 2 (n=3). Less than four hours of etching resulted in blunt needles with strong etching of the lumen of the capillary. Etch times between 4 and 22 hours resulted in increasingly long, thin needles of up to 4 cm. Longer etching resulted in fully dissolving extended portions of the capillary, leaving a short microneedle with an irreproducible tip shape.

The observed lumen shapes are caused by displacement of the silicone oil. At the start of etching a small amount of HF is sucked into the lumen by capillary force, etching a short length from the inside. After 4 hours this portion is fully etched away, leaving only the part of the capillary that was protected from inside etching (figure 2c, h).

The outer shape of the needle can be explained by the interaction between the HF and the polyimide coating. The coating is slightly permeable to HF. When HF permeates through the coating it slowly etches the fused silica and delaminates the coating and the fused silica. When the coating is separated from the capillary, HF is sucked in between by capillary force. This HF results in the strong etching that determines the final tip shape. As HF supply through the coating is small, mass transport by diffusion from the tip is limiting the etch rate, resulting in slower etching further away from the tip. This gradient in etch rate results in the sharp tip of the microneedle. Gluing the microneedles into Luer adapter resulted in needles that can be used and connected with the ease of conventional needles (figure 2).

The relatively short needles obtained by 4 hours of etching were most suitable for injection into the skin because of their rigidity and robustness. While not suitable for injections into the skin, the extremely long needles can be useful for other applications. The ultra-high aspect ratio of these needles (4 cm long, 20 pm tip diameter, figure 2i) could be very useful for deep penetration into soft fragile tissue, e.g. for intracerebral injections.

Applicator

Microneedle applicators are preferably employed or controlled and reproducible microneedle application and reduce the required insertion forces to pierce the stratum corneum for both solid and hollow microneedles.

The microneedle applicator advantageously used in this work was particularly useful for in- and ex-vivo microinjections with easily adjustable insertion speed, depth, angle, and flow rate. Robust performance was observed, i.e. at an insertion speed of 1 m/s no damage of the needles was observed.

In order to construct the system to be completely air free, gas tight syringes and Luer lock connections combined with pressure resistant fused silica tubing were used. This enables constant flow generation at high pressures. As such, the system could be used at relatively high flow rates without using hyaluronidasem, at 2 pL/min, as compared to literature values of single microneedle injections of 50-300 nL/min.

The present sytem comrpsiing the needle sadn applicator was found to to aloe higher flow rates as compared to those disclosed for other systems in the literature, which may have been limited by the maximum applicable pressure in systems using common flexible tubing and different pump types.

The full tunability of the device was found very valuable in a research setting, where varying injections into different skin types and layers are required. Whilst completely compatible with use for intradermal microinjections in in vivo arms of humans, the freestanding setup of the device was also particularly suitable for use in animal studies, as it allows for easy handling of the animals.

It is envisioned that a microneedle applicator for routine use may not need the tunability and flexibility offered by the applicator embodiment disclosed above. Hene, the principles may also be applied in a simplified, hence cheaper, handheld device suitable for large scale immunizations. Determination of microneedle insertion depth

It is important to control the depth of hollow microneedle insertion into the skin, since this defines whether or not the microneedle injection causes pain and/or bleedings. Figure 7a-c depicts micrographs of microneedles after instertion into fluorescently labeled ex vivo human skin with a velocity of 1 m/s. The distance between the fluorescent dye transferred from the skin and the tip of the needle was used to measure penetration depth. Figure 4d shows that the measured microneedle insertion depth was on average 1 10±36 pm (mean±SD, n=4) deeper than the aimed microneedle insertion depth. This bias towards a deeper microneedle insertion than the aimed depth could be due to the pressure that is applied by the microneedle applicator supporting plateau onto the skin, causing the skin to bulge into the opening in the plateau (figure 7G). However, when corrected for this bias, this applicator for hollow microneedles enables us to accurately and reproducibly insert a microneedle into the skin at a chosen depth. The shown depth accuracy is sufficient to selectively inject into the dermis in rats and would allow for even more control in the relatively thick human skin.

Microinjection into skin

Intradermal injections where visualized by injection of 3 μΙ_ fluorescein at a flow speed of 2 L/min into ex vivo human skin. Successful injections were observed at all the injection depths that were tested (100-400 pm). As depicted in figure 7, fluorescence from the injected fluorescein was observed below the surface of the skin. No leakage was observed and diffusion into surrounding tissue was indicated by an increasing area of low fluorescence around the injection site. The negative control of fluorescein dispensed on top of the skin showed fast evaporation of the solvent, causing rapid loss of fluorescence intensity. No significant difference in fluorescence intensity between penetration depths from 100-400 pm was observed in the skin (data not shown).

Immunization study

In order to test the applicability of the hollow microneedles for dermal vaccination, rats were immunized with IPV serotype 1. The resulting antibody titers are shown in figure 6. This figure shows that the serum IgG responses after the (intradermal) microinjections with hollow microneedles at a depth of 300 pm are comparable with those after conventional intramuscular immunization and intradermal immunization, and were comparable to IgG responses described in Westdijk, J., et al., above after a single intramuscular immunization with 5 DU IPV type 1 . Furthermore, a higher dose IPV (15 DU) delivered by a microinjection of 9 μΙ_ led to a significant increase of the IgG response as compared to 5 DU microinjection, but no significant difference between either microneedle injection groups and the intramuscular or intradermal group with 5 DU IPV was observed.

Besides the serum IgG responses, the rat sera were analyzed for VN antibodies, as shown in figure 8. Despite the slightly lower VN titers of rats intradermal^ immunized with a microinjection of 5 DU, compared to conventional intramuscular immunization, this difference was not significant. However, there was a significant difference between 5 DU microinjections and 5 DU intradermal immunization. The lower VN titer of the microinjection might be caused by reduced damage of the skin compared to intradermal injection, where the needle is much thicker, is injected at a greater depth, and a larger volume is injected. This intrinsic effect of less invasive injections might be overcome by adding an adjuvant to the IPV suspension, as described for instance in Bal, S.M., et al., Advances in transcutaneous vaccine delivery: do all ways lead to Rome? Journal of Controlled Release, 2010. 148(3): p. 266-282, and Westdijk, J., et al., above. Furthermore, the immune responses by intradermal injection might be higher than by intramuscular injection after booster immunizations, since intradermal immunization can lead to accelerated booster responses, as described in Nicolas, J.-F. and B. Guy, Intradermal, epidermal and transcutaneous vaccination: from immunology to clinical practice. Expert Review Vaccines, 2008. 7(8): p. 1201 -1214. Therefore, in future experiments the insertion depth (50-1000 pm), the volume effect (3-50 μΙ_), and the booster effect (up to 3 immunizations) should be evaluated. However, a higher dose of IPV by microinjections led to comparable VN titers compared to intradermal and intramuscular immunization. Therefore, these data show the applicability of microneedles to successfully induce VN antibodies against polio-virus.

The experimental results clearly indicate that the needle canulla according to the invention allow to penetrate tissue with much reduced damage as compared to conventional needles. Furthermore, alternative uses such as nanolectrospray injectors are also feasible. Yet further, their use as painless injection needles was demonstrated successfully; which shows that not only can microneedles be fabricated using a cheap and scalable wet etching method, but also their use with a preferred microneedle applicator for successful intradermal microinjections into human and rat skin.

Intradermal microinjection of IPV showed comparable IgG responses to conventional intramuscular injections or intradermal injections. Furthermore, immunization with IPV by microinjection led to the induction of VN antibodies. Etched hollow microneedles can offer pain free, minimally invasive intradermal injections for vaccine delivery.