Astaxanthin is a carotenoid species with the highest antioxidant capability. Its natural resource is very rare. The biosynthesis of astaxanthin from β-carotene includes a hydroxylation step and a ketolation step, for which the corresponding enzymes have been characterized in a few species. However, the sequence of these two reactions is unclear, and may vary with different organisms. In this study, we aimed to elucidate this sequence in Synechocystis, which is an ideal cyanobacterial synthetic biology chassis. We first silenced the endogenous carotene oxygenase gene SyneCrtO to avoid its possible interference in the carotenoid metabolic network. We then introduced the β-carotene ketolase gene from Haematococcus pluvialis (HpBKT) and the CrtZ-type carotene β-hydroxylase gene from Pantoea agglomerans (PaCrtZ) to this δCrtO strain. Our pigment analysis demonstrated that both the endogenous CrtR-type carotene hydroxylase SyneCrtR and HpBKT have the preference to use β-carotene as their substrate for hydroxylation and ketolation reactions to produce zeaxanthin and canthaxanthin, respectively. However, the endogenous SyneCrtR is not able to further catalyze the 3,3′-hydroxylation of canthaxanthin to generate astaxanthin. From our results, a higher accumulation of canthaxanthin and a much lower level of astaxanthin, as confirmed using liquid chromatography–tandem mass spectrometry analysis, were detected in our transgenic BKT+/CrtZ+/δCrtO cells. Therefore, we proposed that the bottleneck for the heterologous production of astaxanthin in Synechocystis might exist at the hydroxylation step, which requires a comprehensive screening or genetic engineering for the corresponding carotene hydroxylase to enable the industrial production of astaxanthin.