Inhibitory effects of copper on bacteria related to the free ion concentration

Cu²⁺ ion determinations were carried out in complex and in inorganic salts-glycerol media, to which increasing amounts of Cu(II) had been added, with the ion-specific Cu(II)-Selectrode. Likewise, complexing capacity of bacterial suspensions was estimated by titration with CuSO₄.Copper-sensitive bacteria, e.g.,Klebsiella aerogenes, were inhibited in their growth and survival in the range of 10^–8–10^–6 M Cu²⁺ ion concentrations. In copper-buffered complex media, high copper loads could be tolerated, as growth proceeded with most of the copper bound to medium components. In low-complexing mineral salts media, in which high Cu²⁺ ion concentrations exist at low copper loads, there was competition of Cu²⁺ for binding sites of the cells. Total allowed copper was then determined by the ratio of copper to biomass.Copper-resistant bacteria could be isolated from a stock solution of CuSO₄, containing 100 ppm Cu(II). They were of thePseudomonas type and showed a much higher tolerance towards Cu²⁺, up to 10^–3 M.     

An electron paramagnetic resonance study of native and modified freeze-dried cupric insulin hexamer.

Cupric insulin was modified by the addition of cross-linking disulphide bridges between hexamers. The electron paramagnetic resonance (EPR) spectrum of this freeze-dried material was compared with that of freeze-dried unmodified cupric insulin containing various amounts of copper and added water. The modified insulin was found to have cupric ion sites magnetically very similar to that of native insulin containing two cupric ions per hexamer. Native hexamer produced in the presence of 2 Cu(II) ions per hexamer gave, after freeze-drying, an EPR spectrum with A_∥^Cu=16.5 mT, g_∥=2.285 and g_⊥=2.059 (site 1). The use of 4 or 6 Cu(II) ions per hexamer resulted in spectra with two components-a major component with the same A_∥^Cu and g_∥ values as the sample containing 2 Cu(II) ions (site 1) and an additional minor component (site 2). These sites have been identified with the analogous zinc binding site within the hexamer formed by three B-10 histidine residues (site 1) [1, 2] and the site formed by the B-1 α-amino and A-17 glutamyl-γ-barboxylic acid functions where excess zinc is bound (site 2) [3, 4]. The addition of water to native hexamer containing 2, 4, or 6 Cu(II) ions resulted in the appearance of three distinct EPR absorptions, one of which had the same parameters as the freeze-dried native insulin containing 2 Cu(II) ions per hexamer (site 1). Two further sites appeared (3 and 4) with the following parameters: A_∥^Cu=15.0 mT, g_∥=2.353, and g_⊥=2.07; A_∥^Cu=16.5 mT, g_∥=2.315, and g_⊥=2.07, respectively.     

COPPER TOXICITY TO SKELETONEMA COSTATUM (BACILLARIOPHYCEAE)1,2

Copper toxicity to Skeletonema costatum (Grev.) Cleve has been studied in batch cultures of chemically defined culture media. The alga is relatively insensitive to cupric ion activity, demonstrating no effect on growth up to (Cu²⁺) = 10^?8.5 M. Cultures inoculated from stationary phase stocks exhibit a prolongation of the lag phase with increasing copper concentrations near and above the point of precipitation of the copper. The toxicity of copper is a function of the silicic acid concentration in the medium. This effect is observed in a range of Si(OH)₄ concentrations (10^?5 M to 10^?4 M) above known values for the saturation of silicon uptake kinetics, thus suggesting an influence of copper on silicate metabolism.     

INTERACTIVE EFFECTS OF COPPER AND SILICIC ACID ON RESTING SPORE FORMATION AND VIABILITY IN A MARINE DIATOM1

The toxic effects of copper on resting spore formation and viability in the marine diatom Chaetoceros protuberans Lauder were determined both with and without silicic acid added to the medium. With silicic acid available, partial inhibition of resting spore formation occurred only at the highest cupric ion activity (pCu 8.6), while the percentage of cells forming spores at pCu's 10.2 and 11.3 was nearly the same as in the controls. Without silicic acid added to the medium, sporulation was completely inhibited at pCu 8.6 and greatly inhibited, at pCu 10.2. At pCu 11.3 and in the controls, the rate of spore formation was less than 50%. The results indicate that the inhibition of resting spore formation by copper is related to the concentration of silicic acid available to cells of C protuberans. This is consistent with previous studies which show that copper toxicity during vegetative growth involves interference with silicification in diatoms and is a Junction of the silicic acid concentration of the medium. Viable resting spores of C. protuberans were still present in cultures following exposure to elevated copper concentrations during a 100-day incubation period. This indicates that resting spores can serve to enhance diatom survival in areas polluted by heavy metals.     

INTERACTIVE EFFECTS OF COPPER AND SILICIC ACID ON RESTING SPORE FORMATION AND VIABILITY IN A MARINE DIATOM1

The toxic effects of copper on resting spore formation and viability in the marine diatom Chaetoceros protuberans Lauder were determined both with and without silicic acid added to the medium. With silicic acid available, partial inhibition of resting spore formation occurred only at the highest cupric ion activity (pCu 8.6), while the percentage of cells forming spores at pCu's 10.2 and 11.3 was nearly the same as in the controls. Without silicic acid added to the medium, sporulation was completely inhibited at pCu 8.6 and greatly inhibited at pCu 10.2. At pCu 11.3 and in the controls, the rate of spore formation was less than 50%. The results indicate that the inhibition of resting spore formation by copper is related to the concentration of silicic acid available to cells of C. protuberans. This is consistent with previous studies which show that copper toxicity during vegetative growth involves interference with silicification in diatoms and is a function of the silicic acid concentration of the medium. Viable resting spores of C. protuberans were still present in cultures following exposure to elevated copper concentrations during a 100-day incubation period. This indicates that resting spores can serve to enhance diatom survival in areas polluted by heavy metals.     

Sublethal effects of cupric ion activity on the phototaxis of three calanoid copepods

Using wavelengths near maximal photosensitivity, phototactic responses of two estuarine calanoid copepods (Acartia tonsa, Acartia hudsonica) and one nearshore, neritic copepod (Temora longicornis) were measured after 24 h exposures to sublethal concentrations of free cupric ions. A nitrilotriacetate-trace metal ion buffer system was used to control the free cupric ion activity (pCu = negative log of the free cupric ion activity), which determines organismic response. All three species exhibited positive phototaxis at pCu 13.0 reported for unpolluted surface sea waters and estuarine waters. As cupric ion activity increased, percent positive phototactic response decreased, indicating a strong sublethal effect of free cupric ions on photobehavior. Changes in photobehavior occurred at cupric ion activities that have been reported for many estuaries and coastal waters near urban and industrialized areas. Temora longicornis was much less phototactically sensitive than the two estuarine species. It also exhibited phototactic sign switching as pCu changed.     

EPR study of a copper center in a single crystal of cytosine monohydrate

Copper is found incorporated into the crystal structure of cytosine monohydrate grown from aqueous solution of commercially available cytosine. Upon ionizing irradiation, the crystals exhibited the electron paramagnetic resonance (EPR) spectra characteristic of Cu(II) complex. Planar coordination bonding to the cupric ion, having three nitrogen atoms and an oxygen as ligands, is interpreted to bridge two cytosine molecules, replacing the two cytosine-cytosine hydrogen bonds present in pure crystals. The EPR signals are much stronger for crystals grown from the solutions to which small amount of copper powder were added.     

Synthesis and solution studies of Cu(II) complexes with pyridine derivatives of iminobisphosphonic acids

New 2-pyridyl, 3-pyridyl and 4-pyridyl derivatives of iminobisphosphonic acid were prepared by addition of tris(trimethylsilyl)phosphite to the corresponding derivatives of pyridineimine-methylphosphonates 3 and subsequent methanolysis of the silylated products 4. Solution studies on the coordination abilities of the ligands have shown that these compounds bind copper(II) ion through the tridentate {N,O,O} mode, where Cu(II) is stabilized by two five-membered chelate rings. The complexes obtained are very stable, with the pCu(II) value above 12, and therefore the ligands can be used as powerful chelating agents for copper ion.     

Copper-dependent inhibition and oxidative inactivation with affinity cleavage of yeast glutathione reductase

Effects of copper on the activity and oxidative inactivation of yeast glutathione reductase were analyzed. Glutathione reductase from yeast was inhibited by cupric ion and more potently by cuprous ion. Copper ion inhibited the enzyme noncompetitively with respect to the substrate GSSG and NADPH. The K_i values of the enzyme for Cu²⁺ and Cu⁺ ion were determined to be 1 and 0.35 μM, respectively. Copper-dependent inactivation of glutathione reductase was also analyzed. Hydrogen peroxide and copper/ascorbate also caused an inactivation with the cleavage of peptide bond of the enzyme. The inactivation/fragmentation of the enzyme was prevented by addition of catalase, suggesting that hydroxyl radical produced through the cuprous ion-dependent reduction of oxygen is responsible for the inactivation/fragmentation of the enzyme. SDS-PAGE and TOF–MS analysis confirmed eight fragments, which were further determined to result from the cleavage of the Met17-Ser18, Asn20-Thr21, Glu251-Gly252, Ser420-Pro421, Pro421-Thr422 bonds of the enzyme by amino-terminal sequencing analysis. Based on the kinetic analysis and no protective effect of the substrates, GSSG and NADPH on the copper-mediated inactivation/fragmentation of the enzyme, copper binds to the sites apart from the substrate-sites, causing the peptide cleavage by hydroxyl radical. Copper-dependent oxidative inactivation/fragmentation of glutathione reductase can explain the prooxidant properties of copper under the in vivo conditions.     

ACCUMULATION OF DOMOIC ACID BY THE COASTAL DIATOM PSEUDO‐NITZSCHIA MULTISERIES: A POSSIBLE COPPER COMPLEXATION STRATEGY

Domoic acid (DA) is a neurotoxic amino acid produced by several members of the diatom genus Pseudo‐nitzschia. Trophic transfer of DA up the food chain has been implicated in the deaths of 100's of marine birds and marine mammals along the central California Coast. The physiological function of DA in Pseudo‐nitzschia spp. has not been defined, although some evidence indicates that elevated metal concentrations can induce DA accumulation (Subba RAO et al., 1998, P.S.Z.N. Mar. Ecol. 19:31). Although California coastal waters have experienced a decline in several heavy metals from 1977–1990, copper concentrations have increased by as much as 25% (Stephenson, M. D. & Leonard, G. H., 1994, Mar. Poll. Bull. 28:148). Many algae produce chelators, including amino acids, in response to toxic [Cu²+] (Wu et al. 1998, J. Phycol. 34: 113). Domoic acid, a tricarboxylic acid, has 4 functional groups that may readily form chelation complexes with transition metals like copper. Copper enrichment experiments indicate that while Cu²+ is toxic to Pseudo‐nitzschia multiseries at total [Cu] greater than 16.1μM (pCu 6.0), intracellular DA accumulation increases up to this point with no decline in growth rates relative to cultures grown in standard enriched seawater. These data suggest that DA may be accumulated by P. multiseries to mitigate the toxicity of elevated [Cu²+]. Chemiluminescence will be used to quantify the binding affinity (expressed as conditional stability constants, Kc) of DA for Cu²+. Defining the Cu‐DA dose response relationship in Pseudo‐nitzschia can facilitate prediction of future toxic bloom events.     

Morphological development and cytochrome c oxidase activity in Streptomyces lividans are dependent on the action of a copper bound Sco protein

Copper has an important role in the life cycle of many streptomycetes, stimulating the developmental switch between vegetative mycelium and aerial hyphae concomitant with the production of antibiotics. In streptomycetes, a gene encoding for a putative Sco-like protein has been identified and is part of an operon that contains two other genes predicted to handle cellular copper. We report on the Sco-like protein from Streptomyces lividans (Sco^Sl) and present a series of experiments that firmly establish a role for Sco^Sl as a copper metallochaperone as opposed to a role as a thiol-disulphide reductase that has been assigned to other bacterial Sco proteins. Under low copper concentrations, a Δsco mutant in S. lividans displays two phenotypes; the development switch between vegetative mycelium and aerial hyphae stalls and cytochrome c oxidase (CcO) activity is significantly decreased. At elevated copper levels, the development and CcO activity in the Δsco mutant are restored to wild-type levels and are thus independent of Sco^Sl. A CcO knockout reveals that morphological development is independent of CcO activity leading us to suggest that Sco^Sl has at least two targets in S. lividans. We establish that one Sco^Sl target is the dinuclear Cu_A domain of CcO and it is the cupric form of Sco^Sl that is functionally active. The mechanism of cupric ion capture by Sco^Sl has been investigated, and an important role for a conserved His residue is identified.     

A first model for the oxidized active form of the active site in galactose oxidase: a free-radical copper complex

 Copper(II) complexes derived from the tripodal ligand bis(3′-t–butyl-2′-hydroxybenzyl)(2-pyridylmethyl)amine (LH₂) have been studied in order to mimic the redox active site of the free radical-containing copper metalloenzyme galactose oxidase. In non-coordinating solvents such as dichloromethane, only an EPR-silent dimeric complex was obtained (L₂Cu₂). The crystal structure of L₂Cu₂ revealed a "butterfly" design of the [Cu(μOR)₂Cu] unit, which is not flattened and leads to a short Cu–Cu distance, the t–butyl groups being localized on the same side of the [Cu(μOR)₂Cu] unit. The dimeric structure was broken down by acetonitrile or by alcohols, leading quantitatively to a brown mononuclear copper(II) complex. UV-visible and EPR data indicated the coordination of the solvent in these mononuclear complexes. Electrochemical as well as chemical (silver acetate) one-electron oxidation of acetonitrile solutions of the monomeric complex led to a yellow-green solution. Based on EPR, UV-visible and resonance Raman spectroscopy, the one-electron oxidation product was identified as a cupric phenoxyl radical system. It slowly decomposes into a product where the ligand has been substituted (dimerization) in the para position of the hydroxyl group, for one of the phenolic groups. The data for the one-electron oxidized species provides strong evidence for a free-radical copper (II) complex. Received: 19 July 1996 / Accepted: 16 October 1996     

Cu(II)-reduction by <Emphasis Type="Italic">Escherichia coli</Emphasis> cells is dependent on respiratory chain components

Copper is both an essential nutrient and a toxic element able to catalyze free radicals formation which damage lipids and proteins. Although the available copper redox species in aerobic environment is Cu(II), proteins that participate in metal homeostasis use Cu(I). With isolated Escherichia coli membranes, we have previously shown that electron flow through the respiratory chain promotes cupric ions reduction by NADH dehydrogenase-2 and quinones. Here, we determined Cu(II)-reductase activity by whole cells using strains deficient in these respiratory chain components. Measurements were done by the appearance of Cu(I) in the supernatants of cells exposed to sub-lethal Cu(II) concentrations. In the absence of quinones, the Cu(II)-reduction rate decreased ~70% in respect to the wild-type strain, while this diminution was about 85% in a strain lacking both NDH-2 and quinones. The decrease was ~10% in the absence of only NDH-2. In addition, we observed that quinone deficient strains failed to grow in media containing either excess or deficiency of copper, as we have described for NDH-2 deficient mutants. Thus, the Cu(II)-reduction by E. coli intact cells is mainly due to quinones and to a lesser extent to NDH-2, in a quinone-independent way. To our knowledge, this is the first in vivo demonstration of the involvement of E. coli respiratory components in the Cu(II)-reductase activity which contributes to the metal homeostasis.     

The effects of copper on the photosynthetic response of <Emphasis Type="Italic">Phaeocystis cordata</Emphasis>

We investigated the effects of limiting (1.96 × 10⁻⁹ mol l⁻¹ total Cu, corresponding to pCu 14.8; where pCu = −log [Cu²⁺]) and toxic Cu concentrations up to 8.0 × 10⁻⁵ mol l⁻¹ total Cu (equivalent to pCu 9.5) on growth rates and photosynthetic activity of exponentially grown Phaeocystis cordata, using batch and semi-continuous cultures. With pulse amplitude modulated (PAM) fluorometry, we determined the photochemical response of P. cordata to the various Cu levels, and showed contrasting results for the batch and semi-continuous cultures. Although maximum photosystem II (PSII) quantum yield (Φ_M) was optimal and constant in the semi-continuous P. cordata, the batch cultures showed a significant decrease in Φ_M with culture age (0–72 h). The EC50 for the batch cultures was higher (2.0 × 10⁻¹⁰ mol l⁻¹, pCu9.7), than that for the semi-continuous cultures (6.3 × 10⁻¹¹ mol l⁻¹, pCu10.2). The semi-continuous cultures exhibited a systematic and linear decrease in Φ_M as Cu levels increased (for [Cu²⁺] < 1.0 × 10⁻¹² mol l⁻¹, pCu12.0), however, no effect of high Cu was observed on their operational PSII quantum yield (Φ′_M). Similarly, semi-continuous cultures exhibited a significant decrease in Φ_M, but not in Φ′_M, because of low-Cu levels. Thus, Cu toxicity and Cu limitation damage the PSII reaction centers, but not the processes downstream of PSII. Quenching mechanisms (NPQ and Q _n) were lower under high Cu relative to the controls, suggesting that toxic Cu impairs photo-protective mechanisms. PAM fluorometry is a sensitive tool for detecting minor physiological variations. However, culturing techniques (batch vs. semi-continuous) and sampling time might account for literature discrepancies on the effects of Cu on PSII. Semi-continuous culturing might be the most adequate technique to investigate Cu effects on PSII photochemistry.     

Copper toxicity: evidence for the conversion of cupric to cuprous copper in vivo under anaerobic conditions.

We have determined the toxicity to cells of Escherichia coli B of cupric copper applied under aerobic and anaerobic conditions in two ways. The growth of cells in liquid medium incorporating cupric copper shows differential inhibition, comparing aerobic and anaerobic conditions--toxicity being greater under anoxia. The growth of colonies upon agar plates incorporating cupric copper does not show such a differential effect. We conclude that colonies on plates are largely anoxic even when incubated aerobically. EPR spectra of cells obtained at various times after application of cupric copper under anoxic conditions indicate the conversion of a considerable proportion of the Cu(II) to a non-paramagnetic species, probably Cu(I). We demonstrate that Cu(I) is more toxic than Cu(II) to cells when applied under anoxic conditions and conclude that the difference in toxicity of Cu(II) applied to cells under aerobic and anaerobic conditions results from the greater extent of reduction of Cu(II) to Cu(I) under anaerobic conditions.     

Analytical determination of apparent stability constants using a copper ion selective electrode

Copper(II) complexes of di-, tri- and tetra peptides with previously published protonation constants were re-investigated using pH and copper ion selective electrode (ISE) potentiometry in conjunction with a modified version of HYPERQUAD computer program. The purpose was to demonstrate the suitability of the ISE approach for the determination of apparent stability constants for copper(II) complexes with ligands for which proton stability constants were not available. The interactions of Cu²⁺ with oligopeptides were also analysed using surface enhanced laser desorption/ionisation time-of-flight mass spectrometry (SELDI-ToF-MS). The results provide an insight into the metal complex species formed, their apparent stabilities under selected conditions and the effect of the relative positions of certain amino acids within the peptide sequence.     

Direct evidence for apo B-100-mediated copper reduction: studies with purified apo B-100 and detection of tryptophanyl radicals

Copper binding to apolipoprotein B-100 (apo B-100) and its reduction by endogenous components of low-density lipoprotein (LDL) represent critical steps in copper-mediated LDL oxidation, where cuprous ion (Cu(I)) generated from cupric ion (Cu(II)) reduction is the real trigger for lipid peroxidation. Although the copper-reducing capacity of the lipid components of LDL has been studied extensively, we developed a model to specifically analyze the potential copper reducing activity of its protein moiety (apo B-100). Apo B-100 was isolated after solubilization and extraction from size exclusion-HPLC purified LDL. We obtained, for the first time, direct evidence for apo B-100-mediated copper reduction in a process that involves protein-derived radical formation. Kinetics of copper reduction by isolated apo B-100 was different from that of LDL, mainly because apo B-100 showed a single phase-exponential kinetic, instead of the already described biphasic kinetics for LDL (namely alpha-tocopherol-dependent and independent phases). While at early time points, the LDL copper reducing activity was higher due to the presence of alpha-tocopherol, at longer time points kinetics of copper reduction was similar in both LDL and apo B-100 samples. Electron paramagnetic resonance studies of either LDL or apo B-100 incubated with Cu(II), in the presence of the spin trap 2-methyl-2-nitroso propane (MNP), indicated the formation of protein-tryptophanyl radicals. Our results supports that apo B-100 plays a critical role in copper-dependent LDL oxidation, due to its lipid-independent-copper reductive ability.     

Reduction of marine phytoplankton reproduction rates by copper and cadmium

The reproduction rates of 38 clones of marine phytoplankton were measured in media in which free cupric ion activity was controlled at different levels using a NTA-cupric ion buffer system. The major trend among species in their resistance to copper toxicity was a phylogenetic one, with cyanobacteria being the most sensitive, diatoms the least sensitive, and coccolithophores and dinoflagellates intermediate in sensitivity. The reproduction rates of most of the cyanobacteria were reduced at cupric ion activities above 10⁻¹² M, while most eukaryotic algae still had maximum reproduction rates at 10⁻¹¹ M. Four species, Emiliana huxleyi (Lohm.) Hay & Mohler, Skeletonema costatum (Grev.) Cleve, Thalassiosira pseudonana (Hustedt) Hasle & Heimdal and Thalassiosira oceanica (Hustedt) Hasle were particularly resistant to copper, being able to reproduce well at the highest cupric ion activities tested, 10^−9.5 M and 10^−9.2 M. There was no major difference, however, between neritic and oceanic species in their sensitivity to copper.The sensitivity of 20 species of marine phytoplankton to free cadmium ion activity was measured in a similar manner using an NTA-cadmium ion buffer system. As observed with copper, the prokaryotic cyanobacteria were the most sensitive to cadmium toxicity, diatoms were the least sensitive, and coccolithophores and dinoflagellates were intermediate. All cyanobacteria tested were dead at a cadmium ion activity of 10^−9.3 M whereas the reproduction rates of most of the eukaryotic algae were not reduced significantly until 10^−8.3 M.Comparison of these data with natural concentrations in sea water implies that cadmium is not an important ecological factor in unpolluted waters but natural copper concentrations may inhibit the reproduction of some phytoplankton species, especially cyanobacteria, in upwelled sea water. Copper may influence the seasonal succession of species.     

Evaluation of copper speciation and water quality factors that affect aqueous copper tasting response

This study determined taste thresholds for copper as its speciation was varied among free cupric ion, complexed cupric ion, and precipitated cupric particles. The impact of copper chemistry on taste is important as copper is added to many beverages and can be present in drinking water as a natural mineral or due to corrosion of copper plumbing. A one-of-five test was used to define thresholds with solutions containing 0.025-8 mg/l Cu (from copper sulfate) in distilled or mineralized water of varying pH. The mineralized water was designed to mimic the composition of a typical tap water. Group thresholds for copper in either distilled-deionized water or mineralized water were not significantly different and ranged from 0.4 to 0.8 mg/l Cu. A difference from control test was used to assess the impact of soluble and particulate copper on taste. Soluble copper species, including free cupric ion and complexed copper species, were readily tasted, while particulate copper was poorly tasted.     

Photooxidation of wetland and riverine dissolved organic matter: altered copper complexation and organic composition

In natural waters, the uptake of transition metals such as copper (Cu) by aquatic biota depends on the activity of the free cupric ion ({Cu²⁺}) rather than on total Cu concentration. Thus, an important ecological function of dissolved organic matter (DOM) in aquatic ecosystems is Cu–DOM complexation, which greatly decreases the {Cu²⁺}. However, Cu bioavailability is greatly modified by source and environmental history of DOM because DOM affinity for Cu varies by orders of magnitude among DOM sources; moreover, DOM is photochemically unstable. During 72-h irradiation experiments at intensities approximating sunlight with DOM from a palustrine wetland and a third-order river, we investigated photooxidative effects on DOM complexation of Cu as well as spectral and chemical changes in DOM that might explain altered Cu complexation. Irradiation decreased Cu complexation by riverine DOM, but unexpectedly increased Cu complexation by wetland DOM, resulting in 150% greater {Cu²⁺} in riverine DOM at the same dissolved organic carbon concentrations. The specific ultraviolet absorption (SUVa) and humic substances tracked photochemical changes in the conditional stability constants of Cu–DOM complexes, suggesting that the aromaticity of DOM influences its affinity for Cu. Carbonyl concentration in ¹³C nuclear magnetic resonance spectra (¹³C-NMR) covaried directly with Cu binding-site densities in DOM. However, no aspect of Cu–DOM complexation consistently covaried with fluorophores (i.e., the fluorescence index) or low molecular weight organic acids. Our results suggest that global increases in UV radiation will affect Cu–DOM complexation and subsequent Cu toxicity depending on light regime as well as DOM source. Handling editor: K. Martens