In this report, the modulation and localization of complement receptors CR1 and CR3 in neutrophils were examined with the use of monoclonal antibodies (mab) directed against these membrane proteins. We first studied complement receptor modulation in a patient with neutrophil-specific granule deficiency. With flow cytometric analysis, we determined that, while N-formyl-methionyl-leucyl-phenylalanine (f-met-leu-phe) (10(-6) M) caused an increase in the binding of both anti-CR1 and anti-CR3 mab to normal neutrophils, the fmet-leu-phe-stimulated neutrophils from our patient increased anti-CR1 binding but decreased anti-CR3 binding. This suggested that CR3, but not CR1, might be associated with specific granules. We next studied receptor modulation in organelle-depleted neutrophil cytoplasts obtained from normal donors. Unlike the specific granule-deficient neutrophils, the normal cytoplasts failed to augment expression of either receptor after stimulation. Immunofluorescence studies of permeabilized polymorphonuclear leukocytes (PMN) revealed considerable internal binding of both anti-CR1 and anti-CR3. In additional studies, phorbol myristate acetate (PMA) was used as a stimulus for receptor modulation in normal neutrophils. Unlike fmet-leu-phe and C5a, PMA elicited a biphasic dose-response curve. High doses of PMA (greater than 0.5 ng/ml) caused a reduction in the magnitude of membrane expression of both CR1 and CR3. In studies designed to localize the internal pool of receptors, we evaluated the binding of 125I-anti-receptor mab to plasma membrane-, specific granule, and azurophilic granule-enriched fractions obtained from sucrose gradient fractionation of disrupted neutrophils. 125I-anti-CR1 mab bound to the membrane-enriched fraction but bound little to either granule-enriched fraction. In contrast, 125I-anti-CR3 mab bound more to the specific granule-enriched fraction than to the plasma membrane-enriched fraction. Azurophilic granules showed no increased anti-CR3 binding. Immunoprecipitation of radiolabeled solubilized subcellular fractions with anti-receptor mab confirmed these findings. CR3 was present in the plasma membrane-, and specific granule-enriched fraction but not in the azurophilic granule-enriched fraction. CR1, however, was present only in the plasma membrane-enriched fraction. These data indicate that there are intracellular pools for both the CR1 and CR3, but the intracellular locations for these pools are distinct. The pool for CR3 co-sediments with specific granules, while the pool for CR1 does not. Nonetheless, a variety of stimulatory agents increase and decrease the membrane expression of both receptors in parallel.