Summary• Perturbations of the source-sink balances were performed in field-grown coffee (Coffea arabica) trees to investigate the possible role of carbohydrates in feedback regulation of photosynthesis.• Four treatments were applied at the whole-plant level: (i) complete defruiting and maintenance of the full leaf area, (ii) the half crop load and full leaf area, (iii) the full crop load and full leaf area and (iv) the full crop load and half leaf area. Sampling and measurements were performed twice during the phase of dry matter accumulation of fruits. Gas exchange, chlorophyll a fluorescence, carbon isotope labelling and steady-state metabolite measurements were assessed in source leaves.• The average rate of net photosynthetic rate (A) and stomatal conductance (g s ) were larger (> 50%), and carbon isotope composition ratio was lower, in trees with a full crop load and half leaf area than in defruited trees, with individuals of the other two treatments showing intermediate values. However, differences in A seem unlikely to have been caused either by photochemical impairments or a direct end-product-mediated feedback down-regulation of photosynthesis.• It is proposed that the decreased A in defruited coffee trees was independent of carbon metabolism and was rather directly related to a lower CO 2 availability coupled to lower g s .

Drought is a major environmental constraint affecting growth and production of coffee. The effects of water supply on growth, biomass allocation, water relations, and gas exchange in two coffee progenies representing drought-tolerant (Siriema) and drought-sensitive (Catucaí) genotypes were compared. They were grown in 12-L pots until 4-months old, when they were submitted to two watering treatments for 60 d: plants receiving either 100% transpired water (control plants) or a fraction (about 40%) of the amount of water transpired by control plants (drought-stressed plants). Under control conditions, Siriema grew faster than Catucaí. Regardless of the watering regimes and progenies, relative growth rate (RGR) was positively correlated both with net assimilation rate (NAR) and long-term water-use efficiency (WUE), but not with differences in biomass allocation. Both progenies responded to drought stress through (i) similar decreases in both RGR and NAR with marginal, if any, changes in allocation; (ii) decreases in leaf water potential, which occurred to a greater extent in Catucaí than in Siriema, even though they have showed similar abilities to adjust osmotically and elastically; (iii) similar reductions in net photosynthesis due mainly to nonstomatal factors; and (iv) decreases in transpiration rate coupled with increased long-term WUE. However, the lower transpiration rate and the higher long-term WUE as found in Siriema relative to Catucaí under control conditions persisted under drought conditions. Overall, the major differences between these progenies were largely associated with differences in plant water use, which was likely related to the improved water status of Siriema. The possible implications of selecting coffee genotypes for high WUE are discussed.

Coffee (Coffea arabica L.) plants were grown in small (3-L), medium (10-L) and large (24-L) pots for 115 or 165 d after transplanting (DAT), which allowed different degrees of root restriction. Effects of altered source : sink ratio were evaluated in order to explore possible stomatal and non-stomatal mechanisms of photosynthetic down-regulation. Increasing root restriction brought about large and general reductions in plant growth associated with a rising root : shoot ratio. Treatments did not affect leaf water potential or leaf nutrient status, with the exception of N content, which dropped significantly with increasing root restriction even though an adequate N supply was available. Photosynthesis was severely reduced when plants were grown in small pots; this was largely associated with non-stomatal factors, such as decreased Rubisco activity. At 165 DAT contents of hexose, sucrose, and amino acids decreased in plants grown in smaller pots, while those of starch and hexose-P increased in plants grown in smaller pots. Photosynthetic rates were negatively correlated with the ratio of hexose to free amino acids, but not with hexose content. Activities of acid invertase, sucrose synthase, sucrose-P synthase, fructose-1,6-bisphosphatase, ADP-glucose pyrophosphorylase, starch phosphorylase, glyceraldehyde-3-P dehydrogenase, PPi : fructose-6-P 1-phosphotransferase and NADP : glyceraldehyde-3-P dehydrogenase all decreased with severe root restriction. Glycerate-3-P : Pi and glucose-6-P : fructose-6-P ratios decreased accordingly. Photosynthetic downregulation was unlikely to have been associated directly with an end-product limitation, but rather with decreases in Rubisco. Such a down-regulation was largely a result of N deficiency caused by growing coffee plants in small pots.

Limitations to photosynthesis were explored in leaves from four canopy positions of field-grown, unshaded coffee (Coffea arabica L.), a tropical tree species classified as shade-obligatory. Overall, compared to shade (lower) leaves, sun (upper) leaves had higher net carbon assimilation rate (A) (4.5 against 2.0 micromol m(-2)s(-1) at most) associated with higher electron transport rate (due to a greater irradiance availability) but unrelated to stomatal and mesophyll conductances, which were similar regardless of leaf position. Neither physiological variable directly involved with photosynthetic carbon gain nor those involved with light capture were able to adjust themselves to match the capacity of the photosynthetic machinery to the light supply. We concluded that: (i) there was no major difference in photosynthetic capacity between sun and shade leaves; (ii) the intrinsic low A in coffee was greatly associated with remarkable low diffusive limitations rather than with biochemical or photochemical constraints; and (iii) morphological (e.g., variations in specific leaf area and leaf inclination) or anatomical plasticity should be of greater acclimative value than physiological plasticity as a mean of coffee leaves to respond to changing irradiance.

No abstract