RESEARCH PAPER A/Ci curve analysis across a range of woody plant species: influence of regression analysis parameters and mesophyll conductance Daniel K. Manter1,* and Julia Kerrigan2 1 USDA Forest Service, PNW Research Station, 3200 Jefferson Way, Corvallis, OR 97331, USA 2 Department of Botany and Plant Pathology, Oregon State University, Corvallis, OR 97331, USA Received 1 July 2004 Accepted 21 July 2004 Abstract The analysis and interpretation of A/Ci curves (net CO2 assimilation rate, A, versus calculated substomatal CO2 concentration, Ci) is dependent upon a number of underlying assumptions. The influence of the Ci value at which the A/Ci curve switches between the Rubisco- and electron transport-limited portions of the curve was examined on A/Ci curve parameter estimates, as well as the effect of mesophyll CO2 conductance (gm) values on estimates of the maximum rate of Rubisco-mediated carboxylation (Vcmax). Based on an analysis using 19 woody species from the Pacific Northwest, significant variationoccurredinthe Ci valuewheretheRubisco-and electron transport-limited portions of the curve intersect (Ci_t),rangingfrom20Pato152Paandaveraging c.71Pa and 37 Pa for conifer and broadleaf species, respect- ively. Significant effects on estimated A/Ci parameters (e.g. Vcmax) may arise when preliminary estimates of Ci_t, necessary for the multiple regression analyses, are set either too high or too low. However, when the appropri- ate threshold is used, a significant relationship between A/Ci and chlorophyll fluorescence estimates of carbox- ylation is achieved. The use of the Vcmax parameter to describe accurately the Rubisco activity from the A/Ci curve analysis is also dependent upon the assumption that Ci is approximately equal to chloroplast CO2 con- centrations (Cc). If leaf mesophyll conductance is low, Cc will be much lower than Ci and will result in an underestimation of Vcmax from A/Ci curves. A large range of mesophyll conductance (gm) values was observed across the 19 species (0.00560.002 to 0.18960.011 mol m22 s21 for Tsuga heterophylla and Quercus garryana, respectively) and, on average, gm was 1.9 times lower for the conifer species (0.05860.017 mol m22 s21 for con- ifers versus 0.11260.020 mol m22 s21 for broadleaves). When this mesophyll limitation was accounted for in Vcmax estimates, considerable variation still existed between species, but the difference in Vcmax between conifer and broadleaf species was reduced from c. 11 lmol m22 s21 to 4 lmol m22 s21. For example, A/Ci curve estimates of Vcmax were 31.266.2and42.264.4 lmolm22 s21, and A/Cc curve estimates were 41.267.1 lmol m22 s21 and 45.064.8 lmol m22 s21, for the conifer and broadleaf species, respectively. Key words: A/Ci curve analysis, CO2 assimilation, mesophyll CO2 conductance, photosynthesis, Rubisco, Vcmax. Introduction A/Ci curve (net CO2 assimilation rate, A, versus calculated internal CO2 concentrations, Ci) analysis has become * To whom correspondence should be addressed. Fax: +1 541 750 7329. E-mail: dmanter@fs.fed.us Abbreviations: A, net CO2 assimilation rate aleaf, leaf absorptance Ca, ambient CO2 concentration Cc, Rubisco catalytic site CO2 concentration Ci, internal CO2 concentration Ci_t, actual Ci value at which the A/Ci curve switches between the Rubisco- and electron transport-limited portions of the curve F 9m, maximal fluorescence upon illumination with a 0.8 s 7000 lmol m��2 s��1 saturation flash Fs, steady-state fluorescence gm, mesophyll CO2 conductance J, electron transport rate Jc electron transport rate devoted to carboxylation measured by chlorophyll fluorescence Jmax, A/Ci curve estimate of the maximum rate of carboxylation limited by electron transport Jo, electron transport rate devoted to oxygenation measured by chlorophyll fluorescence Kc, Michaelis��� Menten coefficient for CO2 binding to Rubisco Ko, Michaelis���Menten coefficient for CO2 binding to Rubisco PPFD, photosynthetic photon flux density Rday, rate of respiration in the presence of light S, specificity of Rubisco for O2/CO2 Vcmax, maximum rate of carboxylation limited by the amount, activity, and kinetics of Rubisco Wc, rate of carboxylation limited by the amount, activity, and kinetics of Rubisco Wj, rate of carboxylation limited by ribulose-1,5- bisphosphate regeneration supported by electron transport Wp, rate of carboxylation limited by triose phosphate availability. Journal of Experimental Botany, Vol. 55, No. 408, �� Society for Experimental Biology 2004 all rights reserved Journal of Experimental Botany, Vol. 55, No. 408, pp. 2581���2588, December 2004 doi:10.1093/jxb/erh260 Advance Access publication 22 October, 2004

a common tool to estimate leaf photosynthesis under a wide variety of experimental conditions (Farquhar et al., 1980 Wullschleger, 1993, and references therein Manter et al., 2000). The response function also represents the mecha- nistic basis behind many plant physiology models (Harley et al., 1992 Manter et al., 2003). While the acquisition of A/Ci response curves is relatively quick and easy to perform, inexpensive (after the initial equipment purchase), and non-destructive, verification of biochemical estimates and analysis assumptions has not been widely tested. According to the Farquhar et al. (1980) model, carbox- ylation rates are limited by one of three processes: (i) the amount, activity, and kinetics of Rubisco (Wc), (ii) the rate of ribulose-1,5-bisphosphate regeneration supported by elec- tron transport (Wj), and (iii) occasionally, triose phosphate availability (Wp). Each of these processes can be described mathematically and is expressed at different Ci values. Determination of leaf photosynthesis and gas exchange via A/Ci curve regression analysis necessitates the a priori designation of a Ci threshold at which the A/Ci curve switches between the Rubisco- and electron transport- limited portions of the curve (Ci_t*). It has been well documented that the Wc process occurs at the lowest Ci values (Farquhar et al., 1980) and common values of Ci_t* used for analysis range from 20���25 Pa (Harley et al., 1992 Wullschleger, 1993). However, the effect of Ci_t* values on A/Ci curve parameter estimates have not been well docu- mented, and based on observations working with conifers (Manter et al., 2000 J Kerrigan and DK Manter, unpub- lished data), it was noted that actual Ci_t values may reach 50 Pa or more and differ markedly between plants. A second inherent assumption in A/Ci curve analysis is that Ci is approximately equal to that of the catalytic site of Rubisco (Cc). However, limitations to mesophyll CO2 conductance (gm), which are not incorporated in A/Ci measurements, may result in a difference between Ci and Cc. Furthermore, of the limited number of plant species from which gm has been quantified, considerable variation has been observed, ranging from c. 25 mmol m��2 s��1to 400 mmol m��2 s��1 (von Caemmerer and Evans, 1991 Lloyd et al., 1992 Loreto et al., 1992 Epron et al., 1995). As a consequence of this difference in Ci and Cc, Epron et al. (1995) showed that estimates of the maximum rate of carboxylation limited by the amount, activity, and kinetics of Rubisco (Vcmax) fromA/Ci curves may be lower than those determined from A/Cc curves (Vcmax_ACc). Finally, much of the variation in Vcmax values across species groups (i.e. broadleaves conifers) (Wullschleger, 1993) may be associated with unaccounted differences in gm and an underestimation of the actual Rubisco activity using the Vcmax parameter, since a similar pattern in gm has also been reported (broadleaves conifers) (Evans et al., 1986 von Caemmerer and Evans, 1991 Lloyd et al., 1992 Loreto et al., 1992 Epron et al., 1995). The purpose of this study was to examine specific parameters involved with leaf photosynthesis and gas exchange measurements to optimize A/Ci analysis and estimates of associated processes. The primary objectives were to (i) determine the influence of Ci_t* values on A/Ci curve parameter estimates and (ii) examine the effect of mesophyll CO2 conductance (gm) values on estimates of the maximum rate of Rubisco- mediated carboxylation (Vcmax) and compare A/Ci and A/ Cc curve estimates. Nineteen woody plant species from the Pacific Northwest were used for measurements, and differ- ences between coniferous and broadleaf species were noted. Materials and methods Plant material One- and two-year-old potted seedlings of various Pacific Northwest species were obtained from local nursery stock in the spring and grown under ambient conditions in an outdoor cold-frame on the Oregon State University campus in Corvallis. Plants were irrigated as needed and fertilized with Osmocote Pro 18-8-8 (Scotts-Sierra Horticultural Products Co., Marysville, OH). The 19 species were Abies concolor (Gordon & Glend.) Lindl. ex Hildebr. (white fir), Abies grandis (Douglas ex D. Don) Lindl. (grand fir), Abies magnifica Andr. Murray (California red fir), Abies procera Rehd. (noble fir), Acer circinatum Pursh (vine maple), Acer macrophyllum Pursh (big leaf maple), Alnus rhombifolia Nutt. (white alder), Alnus rubra Bong. (red alder), Corylus cornuta Marsh (beaked hazel), Larix occidentalis Nutt. (western larch), Pinus lambertiana Douglas (sugar pine), Pinus monticola Douglas ex D. Don (western white pine), Populus trichocarpa3deltoides (hybrid poplar), Pseudotsuga menziesii (Mirb.) Franco (Douglas-fir), Quercus garryana Douglas (Oregon white oak), Quercus rubra L. (northern red oak), Rhodo- dendron macrophyllum D. Don ex G. Don (Pacific rhododendron), Rhododendron occidentale (Torr. & A. Gray) A. Gray (western azalea), and Tsuga heterophylla (Raf.) Sarg. (western hemlock). A/Ci curves and fluorescence Simultaneous measurements of A/Ci response curves and chlorophyll fluorescence were measured on current-year foliage (c. 2 cm2 one- sided projected leaf area) from seedlings (n=2���3) of each species using a Li-Cor 6400 portable photosynthesis system (Open System Vers. 4.0, Li-Cor, Inc., Lincoln, NE) within a 10 d period in July. Cuvette conditions were maintained at a photosynthetic photon flux density (PPFD) of 1600 lmol m��2 s��1, relative humidity 60%, and a leaf temperature of 25 8C. Leaf temperatures were measured directly for broadleaf species and calculated using the energy balance method for conifer species. Ambient CO2 concentration (Ca) in the cuvette was controlled with a CO2 mixer across the series of 30, 20, 10, 40, 50, 60, 80, 100, 160, and 200 Pa, and measurements were recorded after equilibration to a steady state (coefficient of variation 2%). CO2 leakage into and out of the empty cuvette was determined at each reference Ca value and used to correct measured leaf fluxes using the equations provided in the Li-Cor operator���s manual (see also Bernacchi et al., 2002). Following measurements, one-sided projec- ted leaf area for conifers (broadleaf species��� leaves were large enough to fill the entire cuvette) were estimated by placing needles between glass plates and digitally estimating leaf area (Agimage, Decagon Devices, Pullman, WA). Non-linear regression techniques, based on the equations of Farquhar et al. (1980) and later modified by Sharkey (1985) and Harley and Sharkey (1991), were used to estimate Vcmax, Jmax (the maximum rate of carboxylation limited byelectron transport), and Rday (rate of respiration in the presence of light) for each A/Ci curve. In some 2582 Manter and Kerrigan