The Journal of Experimental Biology 202, 3089Ð3099 (1999) 3089 Printed in Great Britain © The Company of Biologists Limited 1999 JEB2272

THE SITES OF RESPIRATORY GAS EXCHANGE IN THE PLANKTONIC MAGNA: AN IN VIVO STUDY EMPLOYING BLOOD HAEMOGLOBIN AS AN INTERNAL OXYGEN PROBE

R. PIROW*, F. WOLLINGER AND R. J. PAUL Institut für Zoophysiologie, Universität Münster, Hindenburgplatz 55, 48143 Münster, Germany *e-mail: [email protected]

Accepted 13 September; published on WWW 28 October 1999

Summary Recent studies on Daphnia magna have revealed that the 6 mmHg (0.8 kPa) for the rostrum. Although not all parts feeding current is important for uptake of oxygen from the of the circulatory system could be analyzed using this ambient medium. Respiratory gas exchange should technique, the data obtained from the accessible regions therefore mainly occur within the filtering chamber, whose suggest that the inner wall of the carapace is a major site boundaries are formed by the trunk and the extended of respiratory gas exchange. Taking the circulatory pattern carapace shell valves. The precise site of gas exchange in and the flow pattern of the medium in the filtering chamber the genus Daphnia is, however, a matter of conjecture. We into consideration, it becomes clear that the haemolymph, have developed a method of imaging the haemoglobin after passing from the limbs to the carapace lacuna, oxygen-saturation in the circulatory system of transparent becomes oxygenated while flowing through the ventral part , which provides an opportunity to localize oxygen of the double-walled carapace in a posterior direction. The uptake from the environment and oxygen release to the laterally flattened rostral region, where sensory and central tissues. Experiments were carried out at 20 ¡C on nervous system structures are located, seems to have direct 2.8Ð3.0 mm long parthenogenetic females maintained in diffusive access to ambient oxygen, which could be hypoxic culturing conditions, which had resulted in an especially advantageous during severe hypoxia when the increased haemoglobin content in the haemolymph. In convective transport systems fail to supply enough oxygen lateral views of D. magna, the highest values of to that region. haemoglobin oxygen-saturation occurred near the posterior margin of the carapace and, surprisingly, in the rostral part of the head. The ambient oxygen partial Key words: Crustacea, , , Daphnia magna, pressures at which haemoglobin was half-oxygenated were gas exchange, respiratory protein, haemoglobin, oxygen-saturation, 15 mmHg (2.0 kPa) for the posterior carapace region and spectral imaging, zooplankton.

Introduction Aerobic energy production depends on the continuous respiratory gas exchange in the water flea Daphnia magna. The exchange of oxygen and carbon dioxide between the cellular literature on crustacean biology (Gerstaecker, 1866Ð1879; combustion sites of an organism and its environment. With Giesbrecht, 1921; Storch, 1925; Krumbach, 1926Ð1927; increasing body size, the metazoans of the higher phyla have Flößner, 1972; Villee et al., 1979; Gruner, 1993) presents had to evolve dedicated organs with surface-enlarged thin different views about the sites of respiratory gas exchange in epithelia mediating the transfer of respiratory gases between the genus Daphnia, including suggestions of (i) gill breathing, the ventilatory and the circulatory systems. Such extensively (ii) intestinal respiration and (iii) integumentary respiration. elaborate structures are usually not present in animals smaller Assigned to the class Branchiopoda, the genus Daphnia than a few millimetres in length (Graham, 1988; Rombough possesses vesicle-like epipodites on its thoracic limbs, which and Ure, 1991; Rombough, 1998), which is not particularly have been repeatedly regarded as gills and have sometimes surprising given their larger surface-to-volume ratios (Krogh, been termed branchial sacs (e.g. Claus, 1876). This assumption 1941). Thus, it is sometimes difficult to ascertain precisely is, so far, consistent with the morphological organization of a what the respiratory organ is or, if it is lacking, to determine typical crustacean, in which the gills derive primarily from whether the whole body surface or only parts of it are evaginations of the limb integument (Barnes, 1969; Gruner, employed for integumentary respiration. 1993). In contrast to the gills of the advanced , The present study aims to determine the favoured sites of however, the epipodites of the genus Daphnia are in no way 3090 R. PIROW, F. WOLLINGER AND R. J. PAUL elaborated with respect to surface area and epithelial thickness which indicates that oxygen is extracted from the feeding (Bernecker, 1909; Fryer, 1991) to enhance the rate of transfer current. of respiratory gases. Although overlain by a cuticle that is very It has repeatedly been suggested that the inner wall of the thin (0.2Ð0.5 µm) relative to that of the rest of the leg (1Ð3 µm), carapace is a major seat of respiratory exchange in the genus the epipodites are lined with an epithelium that is considerably Daphnia (Leydig, 1860; Fryer, 1991). Deriving from an thicker (15Ð20 µm) than ordinary epithelium (3Ð5 µm; integumental fold of the maxillary region (Fryer, 1996), the Kikuchi, 1983). The selective stainability of the epipodites by double-walled carapace consists of two shell valves, which silver salts or vital stains (Fischel, 1908; Gicklhorn, 1925; encase the thorax, abdomen and limbs, thus forming the lateral Gicklhorn and Keller, 1925a), formerly misinterpreted as boundaries of the filtering chamber. Taking into consideration characteristic of respiratory epithelia (e.g. Gicklhorn and the water flow within the filtering chamber (Westheide and Süllman, 1931), points in D. magna, as it does in other Rieger, 1996) and the complex circulatory pattern (Hérouard, crustaceans (Panikkar, 1941; Croghan, 1958), to an 1905; Storch, 1925), it seems very likely that the haemolymph osmoregulatory function, which has more recently been enters into intensive gas exchange with the medium when confirmed ultrastructurally (Kikuchi, 1983). The role of the circulating through the spaces between the inner and outer neck or nuchal organ, in D. magna, a morphological feature walls of the carapace shell valves. Utilizing the presence of restricted to the first instar juvenile (Halcrow, 1982), has to be blood haemoglobin (Hb), a respiratory protein with useful seen in the same functional context (Potts and Durning, 1980; oxygen-sensitive spectral characteristics, a newly developed Halcrow, 1982) rather than linked to respiratory gas exchange spectroscopic imaging technique enabled us to test this (Gicklhorn and Keller, 1925b; Dejdar, 1930). hypothesis experimentally. The striking phenomenon of anal water intake prompted Lereboullet (1850) and later Weismann (1877) to assume that intestinal respiration occurred in the daphniids. Anal water Materials and methods uptake, caused by antiperistaltic movements of the hindgut Animals (Hardy and MacDougall, 1895), was later related to turgor Female water fleas Daphnia magna Straus were cultured restoration (Fox, 1952; Fryer, 1970). It is also thought to under the conditions described previously (Pirow et al., 1999). improve the efficiency of food utilization in the intestine To induce an increased blood haemoglobin (Hb) concentration, (Fryer, 1970). parthenogenetic offspring were raised under conditions of General integumentary respiration seems plausible because moderate hypoxia (30Ð40 % air saturation) produced by of the large surface-to-volume ratio of this millimetre-sized bubbling nitrogen through the culture medium. According to and because of its delicate thin-walled integument Kobayashi and Hoshi (1982), such hypoxic conditions result in (Halcrow, 1976; Dahm, 1977). This hypothesis was a sevenfold elevation of blood Hb concentration (basic level −1 −1 supported by the finding that the beating rate of the thoracic 1gl or 0.06 mmol O2 l ) in 2.5 mm long adult females. The limbs stays constant in D. magna (Heisey and Porter, 1977; animals used in our experiments had a body length ranging from Paul et al., 1997) when the ambient oxygen concentration 2.8 to 3.0 mm, measured from the anterior part of the head to decreases. If the limb movements serve for ventilation, then the posterior edge of the carapace at the base of the apical spine. the expected response to hypoxia in a water-breather with oxyregulatory capacities would be an enhanced limb beating Preparation of animals for experiments rate (Randall et al., 1997). In the oxyregulating D. magna, The experiments were carried out at 20 ¡C in a thermostatted however, systemic responses differ from those of a typical perfusion chamber (see Paul et al., 1997) that allowed water-breather (Paul et al., 1997). The fact that there is no microscopic observation of single animals. To analyze its increase in the limb beating rate need not be regarded as spectral characteristics, the animal was immobilized by glueing negative proof of ventilatory function. In an attempt to filter its apical spine to a 1 cm long synthetic brush-hair (histoacryl out as much food as possible from the ambient medium, when adhesive; B. Braun Melsungen AG, Melsungen, Germany; there is little or no food available, planktonic filter feeders Cowles and Strickler, 1983). The animal was positioned such as D. magna exhibit close to maximum limb beating lateral-side down with the opposite side of the brush-hair and rates (Porter et al., 1982). Elevated food concentrations lower the distal part of the ipsilateral second antenna glued onto a limb beating rate in D. magna and, surprisingly, the expected coverslip. Owing to the curvature of the carapace shell valve, ‘hyperventilatory’ response can then be evoked by reducing the carapace came into contact with the coverslip at the level ambient oxygen concentration (R. Pirow and I. Buchen, in of the base of the middle limbs pairs. The flow of medium preparation), indicating that the limb movements do indeed around the animal was consequently blocked only at this have a ventilatory function. Respiratory gas exchange should contact site and was somewhat reduced at points surrounding therefore occur within the animal’s filtering chamber, this area. The coverslip with the tethered animal was placed because this region is well irrigated with fresh ambient water onto the glass bottom of the perfusion chamber, which was during the steady process of filter feeding. Moreover, the then sealed with a transparent screw-top without touching the oxygen partial pressure was found to be lowered in the animal contralaterally. Experimental animals were perfused medium leaving the filtering chamber (Pirow et al., 1999), from the anterior end with medium of variable oxygen partial Respiratory gas exchange in Daphnia magna 3091 A B

Camera controller y Image ... Ix,y()λ CCD camera data

Acquisition

interface x I0()λ

Computer 400 401 ... 436 437 λ (nm) Objective lens

Specimen D/A

converter λ

stage () x,y A Wavelength control I0()λ Ax,y(λ= ) log 10 —— Light guide Ix,y()λ Monochromator Absorbance, 400Wavelength λ (nm) 437

Fig. 1. (A) Schematic diagram of the spectrophotometric microscope used for haemoglobin imaging. Monochromatic light supplied by a computer-driven monochromator was used for illumination in the transmission mode. The microscopic image was digitized using a 16-bit slow-scan CCD camera and transferred to a computer. (B) A stack of images was taken as the illumination wavelength was increased gradually from 400 to 437 nm in steps of 1 nm. The absorption spectrum Ax,y(λ) of a selected x,y position of the image of the specimen was determined by scanning through the image stack along the wavelength axis. Ax,y(λ) was derived from the spectrum Ix,y(λ) and the reference spectrum I0(λ), which were obtained from a position inside and outside the image of the specimen, respectively.

pressure (PO∑) (see Pirow et al., 1999). Taking into account the 138, Princeton Instruments) and were transferred to the specific systemic adjustments of D. magna in response to computer via a high-speed serial interface (430 kHz maximum changes in PO∑ (Paul et al., 1997), an acclimation period of at pixel rate, Princeton Instruments). The low noise and the large least 10 min preceded the data acquisition at each PO∑ level, dynamic range of the CCD camera allowed the resolution of and this was found to be sufficient for the animal to attain a minor differences in light absorption. new stable heart rate and Hb oxygen-saturation (R. Pirow, The imaging software WinView and WinSpec (Princeton C. Bäumer and R. J. Paul, unpublished data). Instruments) were used for image acquisition and analysis. The built-in C-like programming language was employed to Experimental arrangement for spectral imaging generate macros, which automated image-operation sequences For spectral imaging, a series of gray-scale images of the and synchronized image acquisition and the selection of specimen was acquired while changing the wavelength of illumination wavelength. monochromatic illumination. The apparatus (Fig. 1A) consisted of an inverted microscope (Zeiss Axiovert 100, Carl Details of image acquisition and image analysis Zeiss, Oberkochen, Germany) combined with a computer- For image acquisition, we used an exposure time of 20 ms driven scanning-grating monochromator (T.I.L.L. Photonics, and binning in the range 2×2 to 3×3 pixels. Binning had the Planegg, Germany; 75 W xenon arc lamp, spectral range advantage of reducing acquisition time and saving storage 260Ð680 nm, spectral bandwidth 13 nm, response time <2 ms) capacity at the expense of spatial resolution. To compensate as an illumination system. Collimated monochromatic light for CCD dark charge, a background image, taken with the was guided to the microscope via a quartz fibre-optic light camera shutter closed, was automatically subtracted from the guide (1.5 mm in diameter). The illumination wavelength was incoming image data. Inhomogeneous illumination of the set by a computer equipped with a D/A converter (DAS1602, microscopic field was automatically corrected by the image- Keithley Metrabyte, Taunton, MA, USA). A 16-bit liquid- acquisition programme. Prior to the wavelength scan, the nitrogen-cooled slow-scan CCD camera (576×384 pixels; experimental chamber with the animal inside was placed under LN/CCD-576E, Princeton Instruments, Trenton, NJ, USA) was the microscope. While changing the illumination wavelength mounted on the camera adapter of the microscope for image gradually from 400 to 437 nm in 1 nm steps, a stack of 38 acquisition. Images were digitized by a CCD controller (ST- images was taken within 4Ð11 s. 3092 R. PIROW, F. WOLLINGER AND R. J. PAUL

A B 0.9 Hb fit H ()λ H ()λ o d n = 0.40 0.8 a = 0.910 )

λ = ( b 0.022 A 0.7 r2= 0.994

0.6

Absorbance, 0.5 AaHb()λ=x () λ+

0.4 HnHnHxo()λ= () λ+(1 − ) d () λ

0.9 Gauss fit λ0= 418.8 nm 0.8 c = 0.339 ) A

λ = ( d 13.60 A 0.7 g = 0.432 r2 = 0.995 0.6 Absorbance,

Absorbance, 0.5 2 λ−λ0 1.0 −0.5  0.4 Ac()λ =+e  d  g

400 410 420 430 Wavelength,λ (nm)

Fig. 2. Identification of haemoglobin (Hb) spectra and determination of Hb oxygen-saturation. (A) As an example, an area (18×19 pixel) of the head region of the animal was selected, and the absorbance spectra in the wavelength range 400Ð437 nm were determined. Non-Hb spectra can easily be distinguished from Hb spectra, which feature the Soret band with peak wavelengths ranging from 414 to 427 nm. (B) Two regression equations, Hb fit and Gauss fit, were applied to identify Hb-characteristic absorbance spectra. On the basis of a weighted summation of oxyhaemoglobin (oxy-Hb) and deoxyhaemoglobin (deoxy-Hb) reference spectra, Ho(λ) and Hd(λ), the Hb fit yielded the oxygen-saturation coefficient n, the concentration/path length parameter a and the light-scattering parameter b. The Gauss fit was used to determine the peak 2 wavelength λ0 (arrows). A spectrum was identified as being Hb-characteristic if a, b, λ0 and both correlation coefficients r contained plausible values (see Table 1); otherwise, data were omitted. Oxy- and deoxy-Hb reference spectra, Ho(λ) and Hd(λ), were obtained from a diluted solution of Daphnia magna Hb.

The image stack could be regarded as a three-dimensional Libbert, 1987) in the gut lumen and ingested carotenoids data package of intensity values I(x,y,λ) comprising the (absorption maxima in the range 450Ð500 nm; Herring 1968), intensity (I) for each pixel (x,y) of the image as a function of which can accumulate in the gut wall, the fat cells and the the wavelength (λ). To retrieve the intensity spectrum Ix,y(λ) ovaries (Green, 1957). Although carotenoids are transported in for a selected x,y position, the image stack was scanned along the circulatory system, the haemolymph colour in Hb-rich the wavelength axis (Fig. 1B). The corresponding absorption D. magna results from Hb (Herring, 1968). When spectrum Ax,y(λ) was calculated according to Lambert–Beer’s spectroscopically analyzing haemolymph spaces not law by taking log10{[I0(λ)]/[Ix,y(λ)]}, where I0(λ) is the obstructed by the organs mentioned above, haemolymph Hb is reference spectrum (Fig. 1B). The value for I0(λ) was retrieved identifiable if present at sufficiently high concentration from a region outside the image of the animal and represented (2.5 g l−1; Kobayashi and Takahashi, 1994). This has been light that had not interacted with the specimen. successfully achieved in several studies (Fox, 1948; Green, 1956; Hoshi and Yahagi, 1975; Kobayashi and Takahashi, Identification of Hb spectra and generation of Hb oxygen- 1994). saturation images Advanced theories describe the spectral behaviour of Although D. magna is highly transparent, the presence of chromophores in turbid tissues (Cheong et al., 1990; Seiyama light-absorbing compounds other than Hb must be taken into et al., 1994). However, the transparency of D. magna allowed consideration. The main light absorber in biological fluids and us to choose a less complex approach. Identification of Hb was tissues in the violet and blue parts of the spectrum is the based on a spectral comparison of in vivo spectra with porphyrin system, which is a constitutional part of Hb and reference spectra of oxygenated (oxy-Hb) and deoxygenated cytochromes. Further relevant chromophores in D. magna are (deoxy-Hb) Hb (see below). The regression equation (Fig. 2B; algal chlorophylls (light absorption in the range 400Ð500 nm; Hb fit) was derived from Beer’s law, which describes a two- Respiratory gas exchange in Daphnia magna 3093

Table 1. Empirical data range of regression-analysis (B. Zeis, personal communication) revealed that the peak parameters used for the identification of haemoglobin spectra wavelength of 427 nm was correct. Both the oxy-Hb and Parameter Valid range Description deoxy-Hb spectra were employed as templates with which in vivo Hb spectra were compared (Fig. 2B). a 0Ð1.0 Related to haemoglobin concentration and optical Statistical analyses path length Data are expressed as mean values ± standard deviation b 0Ð1.0 Related to light scattering (S.D.), with N indicating the number of animals examined. λ 0 413Ð428 Peak wavelength Absorption spectra were fitted by linear regression analysis r2 (Hb fit) >0.90 Correlation of determination (Hb fit, see Fig. 2B for regression equation) with stepwise (correlation coefficient) variation of the oxygen-saturation coefficient n from 0 to 1.0. r2 (Gauss fit) >0.90 Correlation of determination To determine peak wavelength, absorption spectra were fitted (correlation coefficient) with a four-parameter Gaussian equation (Gauss fit, Fig. 2B) utilizing the LevenbergÐMarquardt algorithm (Press et al., See Fig. 2 for regression equations. 1992). All regression equations were programmed in the C-like macro language of WinSpec software, which allowed fast off- component system assuming oxy-Hb and deoxy-Hb to be the line analysis of spectral images. only absorbing substances. Incident light can be scattered in D. Haemoglobin oxygen-saturation images were analysed magna by floating haemolymph cells, muscle bundles or further to obtain in vivo oxygen-binding curves for different supporting structures such as the carapace or endoskeletal regions of the circulatory system. Regions of interest were sheets. This effect was taken into account by the scattering selected, each comprising 200Ð400 valid Hb oxygen-saturation factor b (Fig. 2B; Hb fit), which was assumed to be values. For each region of interest, a frequency distribution was wavelength-independent within the narrow wavelength range calculated (class width 2 % Hb oxygen-saturation), and the selected. A second regression equation (Fig. 2B; Gauss fit) was median value was determined. Statistical differences in Hb used to determine the peak wavelength. Regression parameters oxygen-saturation between different haemolymph regions and correlation coefficients were checked for plausibility were assessed using a paired one-sided t-test (P<0.05). (Table 1) to distinguish Hb-characteristic spectra from non-Hb spectra. After analyzing the absorption spectra at all x,y positions, Results the results were depicted as an image (see Fig. 3) in which Images of Hb oxygen-saturation (SO∑) were taken in lateral pixel intensity encoded Hb oxygen-saturation (pseudo-colour views of D. magna at different ambient PO∑ levels (Fig. 3). presentation). Blue represented deoxygenated Hb and red Determination of SO∑ was possible in those haemolymph represented fully oxygenated Hb. For those x,y coordinates spaces that were not obstructed by the large antennae, the gut, where no Hb spectra were detectable, the pixel intensity was the eggs or the beating limbs. The rostral head region and the set to black. posterior parts of the carapace shell valves showed generally higher SO∑ values than the other haemolymph spaces under Preparation of a diluted Hb solution for spectral comparisons normoxic and hypoxic conditions. Because of posture A diluted Hb solution was prepared from haemolymph variations in the abdominal region, SO∑ could not always be samples taken from 20 Hb-rich female adults. After amputating determined in the ventral posterior carapace region the distal half of the second antenna, the oozing haemolymph (Fig. 3A,B). was aspirated into a pulled-glass capillary tube (0.58 mm i.d.), Five positions in the circulatory system of D. magna were which was then emptied into 380 µl of ice-cold phosphate selected for a quantitative comparison. The corresponding in −1 −1 buffer (20 mmol l , pH 7.0) containing 2.63 mmol l vivo oxygen-binding curves showing SO∑ as a function of ascorbate. After centrifugation at 10 000 g (10 min, 4 ¡C), the ambient PO∑ were constructed to make the regional differences supernatant was transferred to a flow-through cuvette (138-OS, in Hb oxygen-saturation more apparent (Fig. 4). At normoxia Hellma, Müllheim/Baden, Germany; 5 mm path length), and (155.7±2.3 mmHg, 20.75±0.31 kPa, N=5), the highest SO∑ absorbance spectra were acquired under normoxic and anoxic values were found in the rostral region (position 1, 90.2±4.8 %) conditions at 20 ¡C using the apparatus described above and in the carapace lacuna near the posterior part of the shell (Fig. 1A). Oxygen was exhausted enzymatically (Lo et al., valves at the base of the apical spine (position 3, 80.4±4.4 %). 1996) by injecting 10 µl of ascorbate oxidase (10 units; Sigma Position 5 yielded the lowest SO∑ (24.1±19.6 %), which Chemical Co.) into the cuvette. The absorbance maxima at represented, in principle, the mean value of the Hb oxygen- 414 nm and 427 nm were identified as those of the Soret bands saturations of two types of overlying haemolymph space at that of oxy-Hb and deoxy-Hb, which corresponded to the values of position: the dorsal lacuna and the carapace lacuna (see Fig. 5). Sugano and Hoshi (1971) for oxy-Hb (414 nm) but not for However, owing to differences in lateral extension and deoxy-Hb (423 nm). However, comparison with the therefore in optical path length, the dorsal lacuna may have absorbance spectrum of purified, deoxygenated D. magna Hb made a greater contribution to this mixed SO∑ value. The 3094 R. PIROW, F. WOLLINGER AND R. J. PAUL

Fig. 3. Images of haemoglobin (Hb) oxygen-saturation of an individual Daphnia magna at various ambient PO∑ levels (A, 153.2 mmHg; B, 15.2 mmHg; C, 5.5 mmHg; D, anoxia). The upper images show the posture of the body within the carapace. Posture variations in the abdominal region were responsible for the different image areas for which valid SO∑ values were obtained (1 mmHg=0.133 kPa).

oxygenation of the haemolymph in the unobstructed carapace (paired one-sided t-test) at two PO∑ levels (155.7 and lacuna at the level of the brood chamber positioned to the right 14.1 mmHg, 20.75 and 1.88 kPa) showed that the SO∑ in the and dorsally from position 5 was generally higher than that rostral head region (position 1) was significantly higher than measured at position 5. It is therefore reasonable to assume that at all other positions (t>3.57, d.f.=4, P<0.05, N=5). Moreover, the SO∑ of the dorsal lacuna, which guides the haemolymph the SO∑ in the carapace lacuna near the base of the apical spine coming from the post-abdomen (see Fig. 5), was actually lower (position 3) was significantly higher than at positions 4 and 5 than that measured at position 5. (t>4.5, d.f.=4, P<0.05, N=5). When entering the extended carapace lacuna, the The in vivo oxygen-binding curves were further used to haemolymph spreads out radially along curved paths that estimate the ambient oxygen partial pressures (P50) at which become confluent at the median dorsal ridge, where blood flow the Hb was half-oxygenated using linear interpolation. For is directed anteriorly to the heart. Located at the dorsal positions 1 and 3, which we regarded as regions containing carapace ridge half-way between position 3 and the heart, oxygenated haemolymph (see Discussion), P50 was 6 mmHg position 4 showed a lower SO∑ (62.8±3.6 %) than position 3. (0.8 kPa) at position 1 and 15 mmHg (2.0 kPa) at position 3. The haemolymph currents in the dorsal lacuna and the carapace lacuna become mixed in the pericardium, from where the haemolymph is aspirated and expelled into the dorsal head Discussion region, where an even lower SO∑ of 57.3±17.6 % was The oxygen-saturation of Hb was measured in the determined (position 2). This hierarchy of SO∑ levels found circulatory system of the planktonic crustacean Daphnia among the five selected haemolymph regions persisted at lower magna with two-dimensional spatial resolution. In contrast to ambient oxygen concentrations. A statistical comparison previous in vivo studies (Fox, 1945; Kobayashi and Tanaka, Respiratory gas exchange in Daphnia magna 3095

100 A 1 80 ) % 60 2

40

20

0 B 100

3 80 ) Hb saturation ( % 4 60

40

Hb saturation ( 5 20

0 0 10 20 30 150 160 Oxygen tension (mmHg)

1

2 Fig. 5. Schematic representation of the main haemolymph spaces and currents in Daphnia magna. The simplified ventral view (A) shows the second and third limb pair; the blood flow of the first and fourth limb pair is only partially indicated. In the dorsal view (B), a piece of the left shell valve at the level of the brood chamber was removed. The ventral (green) and dorsal membranes (yellow) separate the trunk, thereby forming three main blood spaces: the ventral, 4 intestinal and dorsal lacunae. The ventral integument is folded back medianly to the ventral membrane, thus dividing the ventral lacuna 5 symmetrically. In addition, two vertical membranes (blue; A) divide the two resulting ventral lacunae and the limbs into medial and 3 lateral compartments. Leaving the head region in a posterior direction, haemolymph enters either the intestinal lacuna or the medial ventral lacunae. From the medial ventral lacunae, currents Fig. 4. In vivo haemoglobin (Hb) oxygen-binding curves at five project into the five limbs pairs and then enter the lateral ventral positions in Daphnia magna. The P ∑ of the perfusion medium was 0 O lacunae. The haemolymph of the first four limb pairs then passes via mmHg (N=4), 4.3±1.7 mmHg (N=2), 14.1±0.7 mmHg (N=5), the so-called pedicles into the carapace lacuna. Some remaining 19.2±0.5 mmHg (N=2) or 155.7±2.3 mmHg (N=5). Hb oxygen- blood from the medial ventral lacunae and that from the fifth limb saturation is given as a mean value and, if N>2, standard deviation. pair (not shown) joins with that of the intestinal lacuna to perfuse the 1 mmHg=0.133 kPa. abdominal tissues before returning via the dorsal lacuna to the pericardium. In the carapace lacuna, the haemolymph spreads out radially along curved paths and then becomes confluent at the median dorsal ridge of the carapace before returning to the 1991), this imaging technique provided a view of the level of pericardium, where the haemolymph is mixed with that of the dorsal haemolymph oxygenation of various body regions, which can lacuna. The two branches of the feeding current inside the filtering be used to localize the uptake of oxygen from the environment chamber, the subcarapace flow and the median filter flow emerging and the release of oxygen to the tissues. This technique holds from the two posterior interlimb spaces, are indicated in light blue further potential for the visualization of rapidly changing (combined from Hérouard, 1905; Kohlhage, 1994, cited in oxygen distributions in small transparent animals with a time Westheide and Rieger, 1996). 3096 R. PIROW, F. WOLLINGER AND R. J. PAUL resolution within the hundred millisecond range (R. Pirow, F. buffer, 20 ¡C) and reported 50 % oxygenation of Hb at Wollinger, U. Baumeister and R. J. Paul, in preparation). 1.1 mmHg (0.15 kPa) for hypoxia-acclimated (Hb-rich) animals. The level of Hb oxygen-saturation in a section of the Taking into account Fick’s first law of diffusion, it becomes circulatory compartment is, in principle, determined by (i) the clear that the conditions for gas exchange across the inner immediate diffusive loss of oxygen to the tissues, (ii) the carapace wall are indeed favourable. The integument immediate diffusive influx of oxygen from the ambient consists of an outer cuticular and an inner epithelial layer medium, and (iii) the oxygenation level and the flow rate of (Gruner, 1993), whose diffusional resistances to oxygen differ the haemolymph entering that section. Because the different by a factor of approximately 10 (Krogh, 1919). The cuticular sections of the open circulatory system are directionally linked layer therefore represents a crucial diffusional barrier. D. by the blood flow pattern, the corresponding Hb oxygen- magna shows no marked cuticular thickening (Halcrow, 1976; saturation values can be related to each other to characterize Dahm, 1977), and the whole integumental covering has been the diffusive exchange processes for an individual section. found to be permeable to oxygen (R. Pirow, F. Wollinger, U. Although the trunk, the limbs and the ventral half of the Baumeister and R. J. Paul, in preparation). The inner wall of carapace in D. magna were only partly accessible using our the double-walled carapace is covered by a delicate cuticle technique, the data obtained from the remaining body regions (<1.0 µm), which is several times thinner than the cuticle on can provide indications of potential sites of oxygen uptake other parts of the body (Dahm, 1977) thus facilitating gas when the haemolymph and flow patterns in the medium are transfer across the inner carapace wall. taken into consideration. The haemolymph of the first four In addition to the solid integumental barrier, fluid boundary limb pairs passes into the carapace lacuna (Hérouard, 1905; layers represent another obstacle for diffusive gas transport. Fig. 5), where it radiates along curved paths. In the ventral half Millimetre-sized animals live in an environment in which their of the carapace valves, the haemolymph moves in a posterior fluid dynamics can be in the range of low Reynolds numbers direction while coming into close contact with the two where viscous forces are dominant (Koehl and Strickler, 1981; branches of flow of medium inside the filtering chamber (M. Gerritsen et al., 1988). As a consequence, these organisms are Gophen, personal communication; Kohlhage, 1994: cited in barely able to shed the surrounding viscous water layers. The Westheide and Rieger, 1996). The flow of medium and thickness of the boundary layer depends on the velocity of the haemolymph are partly concurrent and partly in cross-current medium relative to the body surface. In Daphnia, reduced orientation to each other. Following the curvature of the boundary layers should occur inside the filtering chamber, ventral-posterior carapace margin, the haemolymph reaches where the medium drains through in a jerky manner the median dorsal ridge of the carapace at the base of the apical accelerated to velocities of 10Ð15 mm s−1 (2Ð3 mm D. spine (position 3, Fig. 4), where we found higher Hb oxygen- magna/pulex; Gerritsen et al., 1988). However, similar saturation values than at positions located downstream of that velocities can occur during swimming, which directly affect region (see Figs 4, 5). The 50 % oxygenation of Hb occurred body surfaces other than those inside the filtering chamber. The at an ambient oxygen partial pressure (P50) of 15 mmHg range of variation is large, from mean swimming speeds of (2.0 kPa). Kobayashi and Tanaka (1991) and Kobayashi and 9Ð15 mm s−1 (2.8 mm for D. magna, 20 ¡C: Kobayashi and Takahashi (1994) measured Hb oxygen-saturation in the dorsal Gonoi, 1985; Fryer, 1991) to sinking speeds of 3 mm s−1 carapace region of hypoxia-acclimated 2.5Ð2.8 mm long D. (1.5 mm D. pulex at 25 ¡C; Gorski and Dodson, 1996), to more magna and reported a P50 of 15Ð17 mmHg (2.0Ð2.3 kPa), stagnant conditions when the animal rests on the bottom with which is consistent with our data. the carapace contacting the substratum. The two latter The high oxygenation level in the posterior part of the situations, as well as conditions in which the animal maintains carapace suggests that influx of oxygen into the haemolymph its position in the water column using strokes of the large occurred during flow through the ventral half of the carapace. antennae, are comparable with our experimental situation. This conclusion is supported by the recent finding that D. magna The maintenance of a large difference in PO∑ is decisive for extracts ambient oxygen from the feeding current (Pirow et al., rapid diffusion of oxygen from the medium to the blood. In the 1999): hypoxia-acclimated females showed a PO∑ of 3.2 mmHg filtering chamber, this requirement is fulfilled by the rapid (0.43 kPa) in the exhalant part of the feeding current under renewal rate of the medium (27Ð87 µl min−1; Pirow et al., hypoxic conditions (16.2 mmHg or 2.16 kPa). Similar hypoxic 1999), which is several times higher than the perfusion rate conditions (15 mmHg or 2.0 kPa) were found to effect 50 % (3Ð4 µl min−1; Paul et al., 1997). The high ventilation-to- oxygenation of Hb in the posterior part of the carapace (this perfusion ratio favours the oxygenation of the haemolymph. In study). As the haemolymph PO∑ in that body region must, in addition, the presence of Hb enhances oxygen transfer from principle, be lower than the PO∑ of the exhalant part of the medium to haemolymph. As long as the Hb is not fully loaded feeding current, thus ensuring the diffusive transfer of oxygen with oxygen, there is only a slight change in blood PO∑ during from the medium to the blood, 50 % oxygenation of Hb should oxygenation, so that the driving force for diffusion can be occur at a haemolymph PO∑ below 3.2 mmHg (0.43 kPa). This largely maintained. inference is reasonable when referring to the data of Kobayashi The presence of the highest oxygenation levels in the rostral et al. (1988), who analyzed the in vitro oxygen-binding region, with a P50 of 6 mmHg (0.8 kPa), was surprising and characteristics of purified D. magna Hb (0.1 mol l−1 phosphate implies a thin-walled rostral integument with a low diffusive Respiratory gas exchange in Daphnia magna 3097 resistance. More important in this context, however, is the shape The technical assistance of Ina Buchen is gratefully of the rostrum, which is flattened laterally. The diffusion acknowledged. We thank Martina Fasel for the excellent distance from the integument to the centre of the haemolymph three-dimensional drawings and G. Sundermann and J. Lange space is consequently very short, thus permitting more rapid for allowing us to inspect their scanning electron microscope penetration of oxygen by diffusion. External diffusive boundary images of water fleas. We are especially grateful to layers adjacent to the rostral integument should be reduced in Alan Rietman Knauth for the linguistic and stylistic free-moving animals, which are propelled forward in a jerky improvements to the manuscript. Supported by the Deutsche manner by powerful strokes of the second antennae. The Forschungsgemeinschaft (Pa 308/7-1). conditions in our experiment were comparable insofar as tethered animals were perfused from the front. Whether there is a substantial contribution to total oxygen uptake is questionable. Nevertheless, this bypass would provide References additional oxygen for sensory and central nervous structures Barnes, R. D. (1969). Zoology. Philadelphia: Saunders. located in the rostral head region (Claus, 1876), which could be Bernecker, A. (1909). Zur Histologie der Respirationsorgane bei of advantage during severe hypoxia when the convective Crustaceen. Zool. Jahrb. 27, 583Ð630. transport system fails to supply enough oxygen to that location. Cheong, W.-F., Prahl, S. A. and Welch, A. J. (1990). A review of the optical properties of biological tissues. IEEE J. Quantum The water flea Daphnia magna shows a variety of Electron. 26, 2166Ð2185. adaptations that have arisen from a filter-feeding life in a Claus, C. (1876). Zur Kenntnis der Organisation und des feineren variable aquatic environment. In addition to what is known Baues der Daphniden und verwandter Cladoceren. Z. Wiss. Zool. about the adjustments occurring at the level of the respiratory 27, 362Ð402. pigment (Kobayashi et al., 1988, 1990) and at the metabolic Cowles, T. J. and Strickler, J. R. (1983). Characterization of feeding and systemic levels (Paul et al., 1997, 1998), this study activity patterns in the planktonic copepod Centropages typicus discloses additional adaptations relevant to respiratory gas Kroyer under various food conditions. Limnol. Oceanogr. 28, transport. The enlarged shell valves of the carapace are not 106Ð115. only a constitutional part of the filter apparatus, forming the Croghan, P. C. (1958). The osmotic and ionic regulation in Artemia lateral boundaries of the filtering chamber, but their special salina (L.). J. Exp. Biol. 35, 219Ð233. structure allows the thin inner wall to be employed for oxygen Dahm, E. (1977). Morphologische Untersuchungen an Cladoceren unter besonderer Berücksichtigung der Ultrastruktur des Carapax. uptake while the thickened outer wall provides stability for the Zool. Jb. Anat. 97, 68Ð126. carapace. The special shape of the rostrum can make ambient Dejdar, E. (1930). Die Korrelationen zwischen Kiemensäckchen und oxygen easily accessible to the central nervous structures, Nackenschild bei Phyllopoden. Z. Wiss. Zool. 136, 422Ð452. which could be important in sustaining vital control functions Fischel, A. (1908). Untersuchungen über vitale Färbung an during hypoxic conditions. Süßwassertieren, insbesondere bei Cladoceren. Int. Rev. Ges. Hydrob. Hydrog. 1, 73Ð141. Flößner, D. (1972). Krebstiere, Crustacea: Kiemen- und Blattfüßer, List of symbols Branchiopoda; Fischläuse, Branchiura. In Die Tierwelt a absorption-related coefficient (haemoglobin Deutschlands (ed. M. Dahl and F. Peus), p. 60. Teil, Jena: Gustav concentration, optical path length) (Hb fit) Fischer. b scattering coefficient (Hb fit) Fox, H. M. (1945). The oxygen affinities of certain invertebrate c, d, g Gauss fit coefficients haemoglobins. J. Exp. Biol. 21, 161Ð165. Fox, H. M. (1948). The haemoglobin of Daphnia. Proc. R. Soc. Lond. A absorbance λ B 135, 195Ð212. Ax,y( ) absorption spectrum at x,y position Fox, H. M. (1952). Anal and oral intake of water by Crustacea. J. λ Ho( ), absorption spectra of oxy-haemoglobin and Exp. Biol. 29, 583Ð599. Hd(λ) deoxy-haemoglobin Fryer, G. (1970). Defaecation in some macrothricid and chydorid Hx(λ) haemoglobin absorption spectra of unknown cladocerans and some problems of water intake and digestion in the oxygen saturation . Zool. J. Linn. Soc. 49, 255Ð270. I intensity Fryer, G. (1991). Functional morphology and the adaptive radiation I0(λ) reference spectrum of the (Branchiopoda: Anomopoda). Phil. Trans. R. Ix,y(λ) intensity spectrum at the x,y position of the image Soc. Lond. B 331, 1Ð99. n haemoglobin oxygen-saturation coefficient Fryer, G. (1996). The carapace of the branchiopod Crustacea. Phil. Trans. R. Soc. Lond. B 351, 1703Ð1712. P50 oxygen partial pressure at which haemoglobin is Gerritsen, J., Porter, K. G. and Strickler, J. R. (1988). Not by half-oxygenated (mmHg, kPa) ∑ sieving alone: Observations of suspension feeding in Daphnia. PO oxygen partial pressure (mmHg, kPa) Mar. Sci. 43, 366Ð376. ∑ SO haemoglobin oxygen-saturation (%) Gerstaecker, A. (1866Ð1879). Die Klassen und Ordnungen der x, y Cartesian coordinates Arthropoden, 5. Band, 1. Abt., Crustacea. Leipzig, Heidelberg: C. λ wavelength (nm) F. Winter’sche Verlagshandlung. λ0 peak wavelength (nm) Gicklhorn, J. (1925). Über spezifische und lokale Reduktion von 3098 R. PIROW, F. WOLLINGER AND R. J. PAUL

Silber- und anderen Metallsalzen in den Kiemensäckchen von Kobayashi, M. and Hoshi, T. (1982). Relationship between the Daphnia M. Lotos 73, 83Ð96. haemoglobin concentration of Daphnia magna and the ambient Gicklhorn, J. and Keller, R. (1925a). Über elektive Vitalfärbungen oxygen concentration. Comp. Biochem. Physiol. 72A, 247Ð249. der Kiemensäckchen von Daphnia magna Müller als Beispiel Kobayashi, M., Nezu, T. and Tanaka, Y. (1990). Hypoxic induction organ- und zellspezifischer Differenzierung. Z. Zellforsch. 2, of hemoglobin synthesis in Daphnia magna. Comp. Biochem. 515Ð537. Physiol. 97A, 513Ð517. Gicklhorn, J. and Keller, R. (1925b). Bau und Funktion des Kobayashi, M. and Takahashi, Y. (1994). In vivo oxygenation of ‘Haftorgans’ von Daphnia bzw. des ‘Kopfschildes’ von hemoglobin in early embryos of Daphnia magna. Comp. Biochem. und auf Grund vitaler Elektivfärbung. Zool. Anz. 64, Physiol. 107A, 127Ð131. 217Ð234. Kobayashi, M. and Tanaka, Y. (1991). Oxygen-transporting function Gicklhorn, J. and Süllman, H. (1931). Die Permeabilität der of hemoglobin in Daphnia magna. Can. J. Zool. 69, 2968Ð2972. Kiemensäckchen von Daphnia magna M. Protoplasma 13, Koehl, M. A. R. and Strickler, J. R. (1981). Copepod feeding 617Ð636. currents: food capture at low Reynolds number. Limnol. Oceanogr. Giesbrecht, W. (1921). Crustacea. In Handbuch der Morphologie der 26, 1062Ð1073. Wirbellosen Tiere, Band 4 (ed. A. Lang), pp. 9Ð252. Jena: Gustav Krogh, A. (1919). The rate of diffusion of gases through animal Fischer. tissues, with some remarks on the coefficient of invasion. J. Gorski, P. R. and Dodson, S. I. (1996). Free-swimming Daphnia Physiol., Lond. 52, 391Ð408. pulex can avoid following Stokes’ law. Limnol. Oceanogr. 41, Krogh, A. (1941). The Comparative Physiology of Respiratory 1815Ð1821. Mechanisms. New York: Dover Publications, Inc. Graham, J. B. (1988). Ecological and evolutionary aspects of Krumbach, T. (1926Ð1927). Handbuch der Zoologie (begr. durch W. integumentary respiration: body size, diffusion and the Kükenthal), 3. Band, 1. Hälfte. Berlin, Leipzig: Walter de Gruyter Invertebrata. Am. Zool. 28, 1031Ð1045. & Co. Green, J. (1956). Variation in the haemoglobin content of Daphnia. Lereboullet, A. (1850). Observations anatomiques et physiologiques. Proc. R. Soc. Lond. B 145, 214Ð232. Mem. Soc. Mus. Hist. Nat. 4, 208Ð212. Green, J. (1957). Carotenoids in Daphnia. Proc. R. Soc. Lond. B 147, Leydig, F. (1860). Naturgeschichte der Daphniden (Crustacea, 392Ð401. Cladocera). Tübingen: Laupp & Siebeck. Gruner, H.-E. (1993). Lehrbuch der speziellen Zoologie (begr. durch Libbert, E. (1987). Lehrbuch der Pflanzenphysiologie. Stuttgart, New A. Kaestner), Band 1, 4. Teil, Wirbellose Tiere: Arthropoda. Jena, York: Gustav Fischer. Stuttgart, New York: Gustav Fischer. Lo, L.-W., Koch, C. J. and Wilson, D. F. (1996). Calibration of Halcrow, K. (1976). The fine structure of the carapace integument of oxygen-dependent quenching of the phosphorescence of Pd-meso- Daphnia magna Straus (Crustacea Branchiopoda). Cell. Tissue Res. tetra (4-carboxyphenyl) porphine: A phosphor with general 169, 267Ð276. application for measuring oxygen concentration in biological Halcrow, K. (1982). Some ultrastructural features of the nuchal organ systems. Analyt. Biochem. 236, 153Ð160. of Daphnia magna Straus (Crustacea: Branchiopoda). Can. J. Zool. Panikkar, N. K. (1941). Osmoregulation in some palaemonid 60, 1257Ð1264. prawns. J. Mar. Biol. Ass. U.K. 25, 317Ð259. Hardy, W. B. and MacDougall, W. (1895). On the structure and Paul, R. J., Colmorgen, M., Hüller, S., Tyroller, F. and Zinkler, function of the alimentary canal of Daphnia. Proc. Camb. Phil. Soc. D. (1997). Circulation and respiratory control in millimetre-sized B 8, 41Ð50. animals (Daphnia magna, Folsomia candida) studied by optical Heisey, D. and Porter, K. G. (1977). The effect of ambient oxygen methods. J. Comp. Physiol. B 167, 399Ð408. concentration on filtering rate and respiration rate of Daphnia Paul, R. J., Colmorgen, M., Pirow, R., Chen, Y.-H. and Tsai, M.- magna mendotae and Daphnia magna. Limnol. Oceanogr. 22, C. (1998). Systemic and metabolic responses in Daphnia magna to 839Ð845. anoxia. Comp. Biochem. Physiol. 120A, 519Ð530. Hérouard, E. (1905). La circulation chez les Daphnies. Mem. Soc. Pirow, R., Wollinger, F. and Paul, R. J. (1999). Importance of the Zool. Fr. 18, 214Ð232. feeding current for oxygen uptake in the water flea Daphnia magna. Herring, P. J. (1968). The carotenoid pigments of Daphnia magna J. Exp. Biol. 202, 553Ð562. Straus. I. The pigments of animals fed Chlorella pyrenoidosa and Porter, K. G., Gerritsen, J. and Orcutt, J. D. (1982). The effect of pure carotinoids. Comp. Biochem. Physiol. 24, 187Ð203. food concentration on swimming patterns, feeding behavior, Hoshi, T. and Yahagi, N. (1975). Studies on physiology and ecology assimilation and respiration by Daphnia. Limnol. Oceanogr. 27, of . XXIX. In vivo equilibrium of oxygen with blood 935Ð949. haemoglobin of Daphnia magna. Sci. Rep. Niigata Univ. Ser. D Potts, W. T. W. and Durning, C. T. (1980). Physiological evolution (Biol.) 12, 1Ð7. in the branchiopods. Comp. Biochem. Physiol. 68A, 475Ð484. Kikuchi, S. (1983). The fine structure of the gill epithelium of a Press, W. H., Teukolsky, S. A., Vetterling, W. T. and Flannery, freshwater flea, Daphnia magna (Crustacea: Phyllopoda) and B. P. (1992). Numerical Recipes in C. Cambridge: Cambridge changes associated with acclimation to various salinities. I. Normal University Press. fine structure. Cell. Tissue Res. 229, 253Ð268. Randall, D., Burggren, W. and French, K. (1997). Eckert Animal Kobayashi, M., Fujiki, M. and Suzuki, T. (1988). Variation in and Physiology: Mechanisms and Adaptations. New York: W. H. oxygen-binding properties of Daphnia magna hemoglobin. Freeman & Company. Physiol. Zool. 61, 415Ð419. Rombough, P. J. (1998). Partitioning of oxygen uptake between gills Kobayashi, M. and Gonoi, H. (1985). Horizontal movement of pale and skin in fish larvae: a novel method for estimating cutaneous and red Daphnia magna in low oxygen concentration. Comp. oxygen uptake. J. Exp. Biol. 201, 1763Ð1769. Biochem. Physiol. 58, 190Ð196. Rombough, P. J. and Ure, D. (1991). Partitioning of oxygen uptake Respiratory gas exchange in Daphnia magna 3099

between cutaneous and branchial surfaces in larval and young blood hemoglobin from fresh-water cladocera, macropa and juvenile chinook salmon Oncorhynchus tshawytscha. Physiol. Daphnia magna. Biochim. Biophys. Acta 229, 349Ð358. Zool. 64, 717Ð727. Vilee, C. A., Walker, W. F. and Barnes, R. D. (1979). Introduction Seiyama, A., Chen, S.-S., Kosaka, H. and Shiga, T. (1994). into Animal Biology. Philadelphia, London, Toronto: Saunders. Microspectroscopic measurement of the optical properties of rat Weismann, A. (1877). Beiträge zur Naturgeschichte der Daphnoiden. liver in the visible region. J. Microsc. 175, 84Ð89. Z. Wiss. Zool. 28, 93Ð254. Storch, O. (1925). Cladocera. In Biologie der Tiere Deutschlands (ed. Westheide, W. and Rieger, R. (1996). Spezielle Zoologie, 1. Teil, P. Schulze), Lief. 15, Teil 14. Berlin: Gebrüder Borntraeger. Einzeller und Wirbellose Tiere. Jena, Stuttgart, New York: Gustav Sugano, H. and Hoshi, T. (1971). Purification and properties of Fischer.