`Antifouling Agents
`
`Mia Dahlstro¨m,1,2 Per R. Jonsson,2 Jukka Lausmaa,3 Thomas Arnebrant,4
`Martin Sjo¨gren,5 Krister Holmberg,6 Lena G.E. Ma˚rtensson,7 Hans Elwing1
`
`1Department of Cell and Molecular Biology, Interface Biophysics, Go¨teborg
`University, P.O. Box 462, SE-405 30 Go¨teborg, Sweden;
`telephone: +46 31 773 2562; fax: +46 31 773 2599;
`e-mail: hans.elwing@gmm.gu.se
`2Department of Marine Ecology, Tja¨rno¨ Marine Biological Laboratory,
`SE-452 96 Stro¨mstad, Sweden
`3Department of Chemistry and Materials Technology, SP Swedish National
`Testing and Research Institute, P.O. Box 857, SE-501 15 Bora˚s, Sweden
`4Surface Chemistry, Biomedical Laboratory Science, Health and Society,
`Malmo¨ University, SE-205 06 Malmo¨, Sweden
`5Department of Medicinal Chemistry, Biomedical Centre, Uppsala University,
`P.O. Box 574, SE-751 23 Uppsala, Sweden
`6Department of Applied Surface Chemistry, Chalmers University of
`Technology, SE-412 96 Go¨teborg, Sweden
`7Department of Zoology, Zoophysiology, Go¨teborg University, P.O. Box 461,
`SE-412 96 Go¨teborg, Sweden
`
`Received 8 July 2003; accepted 15 October 2003
`
`Published online 13 February 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bit.10900
`
`Abstract: In a previous study we found two agents, the
`a2-agonist medetomidine ((F)-4-[1-(2,3-dimethylphen-
`yl)ethyl]-1H-imidazole) and the a2-agonist clonidine (2-
`(2,6-dichloroanilino)-2-imidazoline), that specifically
`and efficiently impede settlement of the barnacle Balanus
`improvisus, one of the most serious biofouling organisms in
`Swedish waters. Medetomidine, but not clonidine, is known
`to adsorb to solid polystyrene (PS) surfaces in the presence
`of salt, a feature that is of particular interest in attempts to
`develop an efficient antifouling surface.We show that
`medetomidine, but not clonidine, has a significant ability
`to adsorb to untreated (hydrophobic) PS in two different
`incubation media: filtered seawater (FSW) and deionized
`water
`(mQ). At negatively charged (hydrophilic) PS,
`medetomidine displays a strong interaction with
`the surface in both incubation media. At the hydrophilic
`PS, clonidine also displays a significant interaction with
`the surface when incubated in mQ and a weaker, but not
`significant, interaction when incubated in FSW. By study-
`ing the effects of time, incubation media, and pH on the
`adsorption of medetomidine and clonidine, we suggest
`that medetomidine is associated to hydrophobic PS by
`means of hydrophobic interactions, while the adsorption
`of medetomidine and clonidine to hydrophilic PS con-
`tains elements of electrostatic interaction. Using time-of-
`flight secondary ion mass spectroscopy (TOF-SIMS) we
`detected only weak signals from medetomidine on the
`hydrophobic PS surfaces, while strong medetomidine
`signals were observed on hydrophilic PS. This suggests
`
`Correspondence to: Hans Elwing
`Contract grant sponsors: Carl Trygger Foundation; Swedish Foundation
`for Strategic Research (SSF); MISTRA; CF Lundstro¨m Foundation; B & B
`Wa˚hlstro¨m Foundation
`
`B 2004 Wiley Periodicals, Inc
`
`that the adsorbed medetomidine, to a greater extent,
`desorbed from the hydrophobic rather than from the
`hydrophilic PS surfaces during exposure to vacuum. The
`strong surface affinity of medetomidine on both types of
`surfaces and the preserved antifouling activity are valu-
`able features in designing a marine coating. B 2004 Wiley
`Periodicals, Inc.
`Keywords: a2 agonists; fouling; barnacle; surface affinity;
`cyprid larvae; TOF-SIMS
`
`INTRODUCTION
`
`Man-made surfaces immersed in the sea will rapidly be
`covered by biological macromolecules such as carbohy-
`drates and proteins, and within weeks, the surface will be
`progressively overgrown by bacteria, algae, and inverte-
`brates in a process called biofouling. Biofouling on the hulls
`of boats and ships causes great economic losses within the
`shipping industry through increased friction, fuel consump-
`tion, and material fatigue. Organisms with calcareous hard
`body armors, such as barnacles and polychaete tubeworms,
`are particularly serious in terms of friction and resistance to
`cleaning (Berntsson, 2001; Harder and Qian, 2000). These
`organisms readily colonize any surface submerged in the
`sea and form dense populations. The standard method to
`combat biofouling is to include antifouling agents in the
`paint formulation. However, several of the toxic substances
`now used for antifouling purposes are facing bans because
`of nontarget effects in the marine environment (reviewed by
`Fent, 1996), and the search for new, nontoxic antifouling
`methods is intense.
`
`Hospira, Exh. 2018, p. 1
`
`
`
`The barnacle Balanus improvisus, which is the target
`organism of this study, has several pelagic naupliar stages
`before molting into the non-feeding settlement stage, the
`cyprid larva. When barnacle cyprid larvae encounter a
`potential settling site, they exhibit a specific surface-bound
`behavior, during which they explore the surface by means
`of their antennae (Berntsson et al., 2000; Crisp, 1974).
`During the exploration of the substratum, the cyprid larva is
`able to perceive and respond to a diverse range of external
`signals. Chemical signals are possibly perceived by several
`aesthetasc sensilla, which are believed to be located on the
`lattice organs situated dorsally on the larvae (Hoeg et al.,
`1998) and also on the antennular segments (Clare and Nott,
`1994; Nott and Foster, 1969). The gregarious settlement in
`like B.
`improvisus,
`many balanomorph barnacles,
`is
`believed to be a result of chemical signals associated with
`the surface and derived both from adult conspecifics and
`from conspecific larvae (Clare et al., 1994; Matsamura
`et al., 1998; Yule and Walker, 1985). Hence, cyprid larvae
`are well adapted for
`receiving signals bound to or
`associated with the surface. A surface-active compound
`in an antifouling paint is thus likely to have a greater
`impact on the settlement of barnacle larvae than a
`compound leaking out of the paint into the water since
`surface activity will increase the concentration close to the
`surface. Compounds with high surface affinities are
`consequently expected to improve the performance of
`antifouling paints with regard to barnacle colonization. We
`have previously described the ability of the two structurally
`similar substances, medetomidine and clonidine, classified
`in vertebrate pharmacology as a2-agonists,
`to inhibit
`settlement of barnacle larvae of B. improvisus in concen-
` 1 to 2.5 ng mL
` 1 when
`trations ranging from 250 pg mL
`present in solution (Dahlstro¨m et al., 2000). Furthermore,
`these substances show large EC50:LC50 ratios (in the 105
`range), implying low toxicity to the larvae. We also found
`that medetomidine, but not clonidine, displayed an obvious
`tendency to accumulate at hydrophilic and hydrophobic
`polystyrene (PS) surfaces, where it was able to exert its
`settlement-inhibiting action (Dahlstro¨m et al., 2000).
`It is an urgent bioengineering challenge to develop a
`coating system that permits slow and controlled release of
`low-toxic, bioactive, antifouling substances. An organic
`antifouling substance, such as medetomidine, is likely to
`interact with several components in the paint film, and
`interaction with the binder can be assumed to be of
`particular importance since the binder is the dominating
`organic component in a paint film. The binder in a marine
`paint formulation is typically a hydrophobic polymer that
`may contain negatively charged carboxylate groups.
`The main objective of the present study is to characterize
`the effect of the antifouling compounds medetomidine and
`clonidine at polymer surfaces. Specifically, we attempted to
`adsorb and desorb medetomidine and clonidine to hydro-
`philic and hydrophobic PS surfaces under various con-
`ditions to characterize the type of surface interactions
`involved. The cyprid settlement assay was used to ensure
`
`that the biological activity of the adsorbed substances was
`preserved, which can be seen as an indirect assessment of
`the presence of the substance at the surface. Furthermore,
`we used time-of-flight-secondary ion mass spectroscopy
`(TOF-SIMS) to chemically detect the adsorbed compounds.
`
`MATERIALS AND METHODS
`
`Larval Rearing and Cyprid Settlement Assays
`
`improvisus were
`the barnacle B.
`Nauplius larvae of
`collected and reared from the brood stock in the culture
`at Tja¨rno¨ Marine Biological Laboratory to yield competent
`cyprid larvae, as described earlier (Berntsson et al., 2000).
`All larvae were used on their first or second day of molting.
`The settlement assays were performed using Petri dishes
`(Ø 48 mm, Nunc) of hydrophilized polystyrene (PS)
`(advancing contact angle with mQ water 81.1j F 1.6j)
`or untreated PS (advancing contact angle with mQ water
`51.4j F 0.6j). These surfaces have previously been
`characterized using X-ray photoelectron spectroscopy
`(XPS), establishing a 12% oxygen content on the hydro-
`philic PS dishes, while on the hydrophobic PS surfaces
`no oxygen was detected (Dahlstro¨m et al., 2000). The
`seawater used was collected via pipes from the Koster fjord
`(depth, 50 m; salinity, 32 F 1x) and then filtered
`through a 0.2-Am filter. Settlement assays were performed
`under static conditions (Dahlstro¨m et al., 2000; Rittschof
`et al., 1984, 1992), and the experiments were maintained
`for 6 – 8 days, after which the dishes were viewed under a
`dissecting microscope to check for attached and metamor-
`phosed individuals and for living cyprids. The chemicals
`used in these experiments were obtained from Sigma
`Aldrich (St. Louis, MO), except for medetomidine, which
`was generously provided by Orion Pharma Oy (A˚ bo,
`Finland). The hydrochloride salts of medetomidine and
`clonidine (Fig. 1A,B) were dissolved in filtered seawater
`(FSW) and diluted to give the desired concentrations.
`
`Figure 1. Chemical configurations of
`clonidine.
`
`(A) medetomidine and (B)
`
`2
`
`BIOTECHNOLOGY AND BIOENGINEERING, VOL. 86, NO. 1, APRIL 5, 2004
`
`Hospira, Exh. 2018, p. 2
`
`
`
`Time-Dependence of Surface Accumulation
`
`A series of experiments were carried out to quantify the
`response of some important factors involved in determining
`the surface accumulation of medetomidine. The mass
`transported from the bulk solution to the solid surface is
`expected to depend on incubation time and incubation
`concentration. Different combinations of time and concen-
`tration were tested using both hydrophilic PS and hydro-
`phobic PS. Dishes of hydrophilic and hydrophobic PS
` 1 or 1 Ag mL
` 1 of
`were filled (20 mL) with 100 ng mL
`medetomidine in FSW and were allowed to stand for 4, 8,
`16, and 32 min to determine the time needed for mede-
`tomidine to accumulate at the surface in sufficient amounts
`to obtain maximum settlement inhibitory activity. After
`incubation, the solution was poured off and each dish was
`washed 4 times with FSW. The dishes were then allowed to
`equilibrate for 2 min with FSW. This water was poured off,
`8 mL of FSW was added together with 20 – 30 cyprids in each
`dish, and the experiment was maintained for 6 – 8 days.
`Dishes of hydrophobic and hydrophilic PS with FSW (8 mL)
`served as controls. Each treatment in either type of dish was
`replicated 3 times, and the experiment was repeated 3 times.
`
`Effect of pH on Surface Accumulation
`
`Medetomidine and clonidine consist of a 1,3-diazacyclo-
`pentadiene and a 1,3-diazacyclopentene ring, respectively,
`connected to a disubstituted benzene ring via a carbon and
`a nitrogen atom, respectively (Fig. 1A,B). The pKa value of
`medetomidine is 7.1, and that of clonidine is around 9; thus,
`medetomidine can be regarded as partially charged
`(f10%) and clonidine to be fully charged (f90%) at the
`pH of seawater (7.9). Two different incubation media were
`used in the experiments to determine the effect of pH on the
`surface interaction of clonidine and medetomidine. FSW
`was used as incubation medium since both clonidine and
`medetomidine have a potential use as additives in marine
`coatings, and mQ was used to establish the importance of
`electrostatic interactions.
`In the first experiment, dishes of hydrophilic and
`hydrophobic PS were incubated with FSW or mQ water
` 1 of medetomidine or clonidine. The
`containing 1 Ag mL
`pH of the FSW was 7.9 F 0.1 (normal pH for seawater),
`5.6 F 0.1 or 3.6 F 0.1, respectively. The mQ water was
`made basic with 0.5 M NaOH (pH 7.9). The FSW and the
`mQ water were made acidic with 1.2 M HCl, and the pH
`was determined with a pH meter (Hanna Instruments,
`Woonsocket, RI). The dishes were incubated for 3 h with
`medetomidine or clonidine. After the incubation, the dishes
`were rinsed and equilibrated as described above and FSW
`(8 mL) was added together with 20 – 30 cyprids. Dishes
`of hydrophobic and hydrophilic PS, which had been treated
`in the manner described above except the addition of mede-
`tomidine or clonidine were filled with FSW (8 mL) and
`served as controls. The experiment was maintained for 6 – 8
`days, after which the dishes were viewed under a dissecting
`
`microscope and the attached and metamorphosed cyprids
`were counted. The treatments were replicated 4 times, and
`each experiment was performed 3 times.
`In a second experiment, the rinsing water was made
`acidic to establish whether the surface bound medetomidine
`could be desorbed from the surface. Dishes of hydrophilic
` 1 of
`and hydrophobic PS were incubated with 1 Ag mL
`medetomidine in FSW (pH 7.9) for 2 h, after which they
`were rinsed 4 times and allowed to equilibrate for 2 min
`with FSW of pH 7.9, 5.0, 4.0, or 3.0. The rinsing water was
`made acidic, and the pH was determined in the same
`manner as described above. FSW (8 mL) was added to each
`dish together with 20 – 30 cyprids, and the experiment was
`maintained for 6 – 8 days. Dishes of hydrophobic and
`hydrophilic PS, which had been treated in the manner
`described above except the addition of medetomidine, were
`filled with FSW (8 mL) and served as controls. The
`treatments were replicated 3 times, and each experiment
`was performed twice.
`
`Analysis of Surface-Bound Medetomidine Using
`Time-of-Flight Secondary Ion Mass Spectroscopy
`(TOF-SIMS)
`
`Hydrophobic and hydrophilic polystyrene samples were
` 1, 30 min at
`incubated in medetomidine solutions (1 Ag mL
`RT) and then analyzed with time-of-flight secondary ion
`mass spectrometry (TOF-SIMS). Non-incubated samples
`of the corresponding PS materials were also analyzed for
`comparison. Positive and negative secondary ion
`mass spectra were acquired from at least two areas of
`500 500 Am on each sample; 25-keV Ga+ ions were used
`as the primary ions at a beam current of 0.5 pA. The spectra
`were acquired under static SIMS conditions,
`i.e.,
`the
`primary ion dose was kept below 1 1013 ions cm
` 2 in
`order to avoid contributions from ion beam damaged areas
`in the spectra.
`
`Statistical Methods
`Results are generally reported as means F standard error.
`Effects of the studied substances on cyprid settlement were
`tested using one- and two-factor analysis of variance
`(ANOVA) with all tested treatments as fixed factors (Winer
`et al., 1991). In all tests an a of 0.05 was used.
`
`RESULTS AND DISCUSSION
`
`To make an in-depth investigation of the factors involved
`in the surface binding of medetomidine and clonidine
`(Fig. 1A,B), we used two types of surfaces: one polar and
`one nonpolar. Previous work has shown that clonidine, in
`solution, displays approximately the same efficacy in
`settlement inhibition as medetomidine but appears to lack
`sufficient surface affinity (Dahlstro¨m et al., 2000). The
`approaches used to determine the mechanisms involved in
`
`DAHLSTRO¨M ET AL.: SURFACE CHEMISTRY OF A BARNACLE-REPELLING COMPOUND
`
`3
`
`Hospira, Exh. 2018, p. 3
`
`
`
`the surface interaction of the antifouling substances were to
`study how time and concentration, as well as pH and type of
`incubation medium, affected the ability of the substances to
`impede cyprid settlement. In addition to the biological assay,
`using cyprid settlement as an indicator of the presence of the
`substance at the surface, we used the surface sensitive
`analytical method TOF-SIMS.
`Figure 2 shows the cyprid settlement inhibition effect of
`medetomidine at different times and concentrations. We
`found that in the hydrophobic dishes incubated with 100 ng
` 1 or 1 Ag mL
` 1 of medetomidine settlement of cyprid
`mL
`larvae was inhibited in a time-dependent manner (2-factor
`ANOVA, F4,20 = 6.7, P < 0.05) (Fig. 2A). This time
`dependence occurred regardless of the concentration used,
`and hence, there was no significant interaction between time
`and concentration (F4,20 = 0.2, P > 0.05). In the hydrophilic
`dishes, settlement was completely inhibited for both
`concentrations already at the incubation time of 4 min
`(two-factor ANOVA, F4,20 = 79, P < 0.05) (Fig. 2B). The
`experiments were conducted three times with the same
`results. We suggest that the much more pronounced effect
`obtained in the hydrophilic dishes is an effect of stronger
`affinity due to electrostatic interactions between medetomi-
`dine and this surface. Medetomidine, which is a weak base,
`will be partially positively charged at the pH used (7.9), thus
`binding can take place through ionic interactions with
`negatively charged carboxylate groups present at the surface
`of the hydrophilic dish (Dahlstro¨m et al., 2000). Hydro-
`
`Figure 2. Time-dependence of the surface accumulation of two different
`concentrations of medetomidine: (A) hydrophobic surfaces; (B) hydro-
`philic surfaces. The response variables of percentage settled are shown as
`means F SE (n = 3). (Note the difference in scale between the two.)
`
`phobic interactions are also likely to contribute to the
`binding to the hydrophilic dish.
`In the binding of
`medetomidine to the surface of the hydrophobic dish, no
`Coulombic interaction forces are involved. The interaction
`must be due to hydrophobic interactions and possibly k –k
`interactions, primarily between the diazacyclopentadiene
`ring as electron-rich k component and surface benzene rings
`as electron-poor k components. Also, k – cation interactions
`involving the partially protonated amino group of medeto-
`midine as the cation component and surface benzene rings as
`the k components may play a role. We conclude by this
`experiment
`that
`the electrostatic binding is the most
`important means of the molecule to interact with the solid
`surface; however, elements of hydrophobic interactions are
`also contributing to the surface affinity of medetomidine.
`When incubating medetomidine and clonidine in mQ at
`three different pH values, we found that medetomidine
`significantly interacted with both types of surfaces in all pH
`treatments tested. Thus, regardless of incubation pH, the
`settlement of cyprid larvae was completely inhibited and no
`significant
`interaction between the factors treatment
`(control vs. medetomidine), and pH was obtained at neither
`the hydrophilic (ANOVA, F2,18 = 0.42, P > 0.05) (Fig. 3A)
`nor the hydrophobic (ANOVA, F2,18 = 0.35, P > 0.05)
`(Fig 3B) surfaces. The results from the incubation of
`clonidine in mQ at different pH displayed the same pattern
`in the hydrophilic dishes (ANOVA, F2,18 = 1.56, P > 0.05)
`(Fig. 3A). In all tests, the factor treatment was significant.
`However, clonidine was still unable to interact with the
`hydrophobic surfaces when incubated in mQ in all the
`different pH treatments, and thus the settlement in these
`dishes equaled that in the control dishes (Fig. 3B).
`The pH strongly affected the ability of medetomidine,
`but not of clonidine,
`to interact with the hydrophobic
`surfaces when incubated in FSW. Our results show that
`medetomidine completely inhibited settlement on hydro-
`phobic surfaces at pH 7.9 and 5.6 (Fig. 3C). However,
`settlement was not inhibited at pH 3.6 (Fig. 3C). These
`effects are detected as a significant
`interaction among
`the factors pH and treatment (control vs. medetomidine)
`(two-factor ANOVA, F2,18 = 54.2, P < 0.05). The major
`contribution to this interaction is the high settlement in the
`dishes incubated with medetomidine in FSW at pH 3.6. On
`the hydrophilic surfaces, medetomidine inhibited settle-
`ment at all three incubation pH values (Fig. 3D). Thus, it
`seems that the pH did not affect
`the ability of mede-
`tomidine to adsorb to the surface (two-factor ANOVA,
`F2,12 = 1.53, P > 0.05). In the dishes incubated with
`clonidine,
`the effect of pH was small both on the
`hydrophilic and on the hydrophobic surfaces (Fig. 3C,D).
`In these experiments, one striking result is the inability
`of medetomidine to interact with the hydrophobic surface
`when incubated in FSW at pH 3.6. This inability cannot
`solely be explained in light of intermolecular repulsion,
`since the molecule is fully charged also at pH 5.6.
`Furthermore, in the absence of salts, medetomidine is able
`to interact with the hydrophobic surface regardless of pH.
`
`4
`
`BIOTECHNOLOGY AND BIOENGINEERING, VOL. 86, NO. 1, APRIL 5, 2004
`
`Hospira, Exh. 2018, p. 4
`
`
`
`Figure 3. pH dependence of the surface accumulation of medetomidine and clonidine. (A) Hydrophobic surfaces, incubation in mQ; (B) hydrophilic
`surfaces, incubation in mQ; (C) hydrophobic surfaces, incubation in FSW; (D) hydrophilic surfaces, incubation in FSW. The response variable is
`percentage settled in control dishes, medetomidine-treated dishes, and clonidine-treated dishes shown as means F SE (n = 4).
`
`We therefore suggest that, when using FSW as incubation
`medium, the positive charge of medetomidine at pH 3.6 in
`combination with the high ionic strength of the FSW alters
`the surface binding behavior to the hydrophobic surfaces. It
`may be speculated that the presence of salts may increase
`self-association, thereby reducing interface accumulation.
`We have made analogous observations concerning the
`adsorption of medetomidine at different pH values and in
`the absence and presence (artificial seawater) of salt in
`experiments using ellipsometry, an optical surface-sensitive
`method whereby the concentration of medetomidine at the
`surface can be directly quantified (Arnebrant et al.,
`unpublished). However, we are not yet able to account
`for the inability of medetomidine to interact with the
`hydrophobic surface at pH 3.6. Clonidine, on the other
`hand, seems unable to sufficiently interact with the
`hydrophobic surfaces, regardless of pH and incubation
`media. Thus, we propose that the interaction of medetomi-
`dine to hydrophobic surfaces contains substantial elements
`of hydrophobic interactions.
`
`The association of medetomidine and clonidine mole-
`cules to the hydrophilic surfaces in the absence of salts is
`probably mediated by electrostatic interactions to nega-
`tively charged carboxylate groups. However, when the pH
`is lowered and thus, the solubility of the molecules is
`increased while the surface is also rendered noncharged,
`the molecules are still able to interact with the surface. It is
`likely, though, that the amount of substance associated to
`the surface decreases with lowered pH as a consequence of
`intermolecular repulsion, a feature that is not detectable by
`the cyprid larvae settlement assay.
`The results of the experiments where the dishes were
`incubated at pH 7.9 but washed with FSW at low pH
`(pH 5.0, 4.0, and 3.0) show that desorbtion of
`the
`medetomidine molecule is not possible by employing these
`washing procedures, regardless of surface and pH. In these
`experiments, the pH of the washing water was not able to
`affect the medetomidine already associated to the surface in
`any of the dishes (one-factor ANOVA, F7,16 = 310.7,
`P < 0.05) (Fig. 4A,B).
`
`DAHLSTRO¨M ET AL.: SURFACE CHEMISTRY OF A BARNACLE-REPELLING COMPOUND
`
`5
`
`Hospira, Exh. 2018, p. 5
`
`
`
`Therefore, we tried in an earlier experiment to quantify the
`adsorbed medetomidine at the PS surface with the use of
`X-ray photoelectron spectroscopy (XPS). However, it was
`not possible to detect the molecule with the use of this
`method, as evidenced by the lack of nitrogen signals. In this
`study, we therefore used TOF-SIMS, which is a more
`surface-sensitive method than XPS and also makes possible
`more direct chemical identification of the molecules in
`question (reviewed by Benninghoven, 1994). Positive
`secondary ion mass spectra from hydrophilic PS surfaces
`exposed to medetomidine and then rinsed with mQ water
`showed a strong signal from protonated medetomidine
`molecular ions (M+H)+ at m/z = 201.1 (Fig. 5A,B, upper
`panels). The strong signal at 95.1 is due to C5H7N2 (Fig. 5C),
`which is a characteristic fragment of the medetomidine
`molecule, as determined from analyses of medetomidine
`adsorbed or deposited as bulk powder on silver performed
`prior to the TOF-SIMS measurements of the surface-
`associated medetomidine. Although the survey spectrum
`was typical of polystyrene (Fig. 5A), clear (M+H)+ and
`C5H7N2 signals from medetomidine could also be detected
`on the hydrophobic PS surfaces exposed on medetomidine
`(Fig. 5B,C). The medetomidine signals were, however, a
`factor of 1,000 lower than for the hydrophilic PS. This result
`does not necessarily mean that the surface coverage of
`medetomidine after adsorption from solution was low. A
`
`Figure 4. Effects of pH of the rinsing water on the surface retained
`medetomidine: (A) hydrophilic dishes; (B) hydrophobic dishes. The
`response variable is percentage settled shown as means F SE (n = 3).
`
`We conclude from these adsorption and desorption
`experiments that hydrophobic interactions constitute the
`dominating attractive interaction at the hydrophobic PS
`surfaces, while for the hydrophilic PS surfaces, electrostatic
`interactions dominate.
`The cyprid settlement assay is a good indicator of the
`presence of the molecule at the surface, but nevertheless, it
`is not a satisfactory tool to actually determine the physical
`presence of medetomidine or clonidine at the PS surface.
`
`Figure 5. Positive secondary ion mass spectra from polystyrene surfaces exposed to medetomidine. (A) Survey spectrum (25 – 210 D); (B) high-
`resolution spectrum of the C5H7N2 fragment ion; (C) high-resolution spectrum from protonated medetomidine molecular ion (M+H)+.
`
`6
`
`BIOTECHNOLOGY AND BIOENGINEERING, VOL. 86, NO. 1, APRIL 5, 2004
`
`Hospira, Exh. 2018, p. 6
`
`
`
`the
`the majority of
`more likely explanation is that
`medetomidine molecules desorbed from the hydrophobic
`PS surfaces prior to analysis (which is done in vacuum). The
`amount of material desorbed may depend on the type of
`interaction to the surface. It seems that the bonding to the
`hydrophilic PS (and also to the silver surface) is strong
`enough to withstand drying and vacuum conditions,
`while in the case of hydrophobic PS the attachment is wea-
`ker (since it is dominated by hydrophobic interactions, as
`discussed below).
`In the TOF-SIMS analysis of the PS surfaces, an
`additional observation was made concerning the effect
`of medetomidine on the hydrophobic PS surfaces. Hydro-
`phobic samples incubated in mQ water showed spectra
`which were not characteristic of pure PS but which contained
`strong signals from saturated hydrocarbon species, indicat-
`ing fatty acid contamination. After incubation in medeto-
`midine solution, these surfaces showed more characteristic
`PS spectra (Fig. 5A, lower panel). Our interpretation of this
`result
`is that
`incubation with medetomidine results in
`removal of impurities from the PS surface, also indicating
`a detergent effect of the molecule.
`Current research on alternatives to toxic additives in
`marine paints is focused on target-specific molecules with
`the aim to reduce environmental impact. Many recent dis-
`coveries of substances acting as potent settlement inhibitors
`include natural compounds with a specific ecological
`function, e.g., furanones from Delisea pulchra (De Nys
`et al., 1995), and steroids from marine sponges and
`octocorals (Kon-Ya et al., 1994, 1997; Rittschof, 1985). In
`addition, some synthetic compounds used in neurological
`research on vertebrates, e.g., catecholamine inhibitors, have
`been found to show strong inhibitory effects on settling
`larvae (Dahlstro¨m et al., 2000; Yamamoto et al., 1998). Any
`substance released into the marine environment is subject to
`a number of processes like transformation, degradation,
`sedimentation, biological uptake, and bioaccumulation,
`which determine its environmental fate (reviewed by
`Thomas, 2000). This fate is largely unknown for most of
`the synthetic substances found to have an antifouling
`activity. In order to improve the performance and to reduce
`the effect on the environment, it is important to have proper
`control of the release of the antifouling substance from the
`paint film.
`The lack of a surface signal of clonidine at the hydro
`phobic surfaces in the biological assay using the cyprid
`larvae, notwithstanding its molecular
`resemblance to
`medetomidine, can in part be explained by the lack of
`hydrophobic side groups such as methyl or ethyl sub-
`stituents that can participate in hydrophobic interactions.
`Thus, even if an interfacial accumulation of clonidine takes
`place at the hydrophobic surfaces, this accumulation is
`apparently not able to withstand extensive rinsing.
`The concept of surface (interface) affinity of antifouling
`molecules may be of particular importance where the cyprid
`larvae are concerned. During the initial exploratory phase,
`the larva investigates the surface with the use of its paired
`
`antennules (Berntsson et al., 2000; Crisp, 1974) in a behavior
`termed ‘‘straight-line walking’’ where cyprids employ their
`antennae in a bipedal walking fashion that can last for as long
`as 8 min without detachment (Lagersson and Hoeg, 2002).
`During this walk, the cyprid repeatedly flicks with the fourth
`antennular segment, which is furnished with chemo- and
`mechanoreceptor organs (Clare and Nott, 1994; Nott and
`Foster, 1969), and thus, the possibilities to sense the presence
`of molecules at the surface with both the third antennular
`segment (the antennular disc) and the fourth segment are
`numerous. We suggest that during the straight-line walking,
`cyprid larvae will encounter and accumulate the surface
`associated medetomidine. As the antennular discs make
`substratum contact, medetomidine will desorb from the
`surface and enter the larvae through the discs, thereby
`impeding the larvae from proceeding to the close inspection
`phase of the settlement process, a process that inevitably
`precedes permanent attachment (Berntsson et al., 2000;
`Crisp, 1974; Lagersson and Hoeg, 2002). Another relevant
`aspect of surface affinity is the degree of affinity of the
`antifouling substance, e.g., medetomidine, for the solid
`surface. If the association of medetomidine to the surface is
`too firm, the molecule may not be released and thus cannot
`reach the target organs of the cyprid larvae. On the other
`hand, if the affinity of the molecule is too low, the high
`interfacial concentration of the molecule, and hence the
`antifouling effect, is lost. Optimally, the degree of affinity
`ought
`to be such as to guarantee a high interfacial
`concentration and at the same time allow for a release of
`the molecule from the surface when cyprid larvae make
`substratum contact.
`The lengthy time of the experiments (6 – 8 days) may
`cause a fraction of the adsorbed medetomidine to desorb
`from the surface into the limited volume (8 mL) of FSW
`used in the Petri dishes. It cannot be excluded that the
`potentially desorbed medetomidine also affects the settle-
`ment response. On the other hand, several observations
`imply that the medetomidine, once adsorbed, is not likely
`to spontaneously desorb from the surface in amounts large
`enough to affect the cyprid larvae. A more detailed study of
`the adsorption and potential desorption kinetics of mede-
`tomidine to solid surfaces is under way. This study will
`employ ellipsometry for quantification of the adsorption of
`medetomidine to well-characterized model surfaces.
`
`CONCLUSIONS
`
`Based on the results obtained in this study, we conclude
`that the high surface affinity of medetomidine is media-
`ted through ionic and hydrophobic interactions, which
`are strong enough to withstand rinsing but weak enough to
`allow a release of the compound from the surface as it
`is encountered by the cyprid larvae. We suggest
`that
`medetomidine would be the better candidate as an
`antifouling agent, largely due to its high degree of surface
`affinity, which implicates possibilities of long-term per-
`formance in the field.
`
`DAHLSTRO¨M ET AL.: SURFACE CHEMISTRY OF A BARNACLE-REPELLING COMPOUND
`
`7
`
`Hospira, Exh. 2018, p. 7
`
`
`
`Our opinion concerning the importance of surface
`affinity of small molecules to polymeric surfaces of use
`in the marine environment has been strengthened by
`observations made in this study, largely due to the unique
`and sensitive in vivo measurements of biologically active
`molecules at the surface, using the settlement assay of
`barnacle larvae.
`
`We acknowledge Martin Ogemark at Tja¨rno¨ Marin