A possible association between depression and either the severity of constipation or dysosmia in Parkinson’s disease (PD) patients was investigated in this cross-sectional study. One-hundred six patients who had the history of PD for less than 5 years were recruited. Depression was measured using the Beck Depression Inventory-II (BDI-II), and our patients were divided into depressive and non-depressive groups (DP: BDI-II ≥ 14; n = 22 and NDP: BDI-II < 14; n = 84). Olfactory dysfunction was assessed by the University of Pennsylvania Smell Identification Test (UPSIT). Constipation severity was defined by stool softener dosage and amount. Statistical analyses with one-tailed T- or chi-squared test, odds ratios (OR), and beta-coefficient were used to determine significant differences. Total scores based on the Unified Parkinson’s Disease Rating Scale (UPDRS) were significantly higher in the DP group. A significant relationship was observed between PD patients with depression and severe constipation; PD patients with depression were more likely to present with severe constipation (OR 5.81; 95% CI 1.24–27.29, p = 0.026, adjusted for age and gender); but the significance became marginal after adjusted for age, gender and UPDRS part 3 (OR 4.46, 95% CI 0.93–21.33; p = 0.061). However, no association between olfactory dysfunction and depression was detected. There were significant positive correlations between BDI-II scores and severe constipation (β ± SE 7.65 ± 2.02; p = < 0.001, adjusted for age and gender; β ± SE 7.06 ± 2.04; p = 0.001, adjusted for age, gender, and UPDRS-3). Besides, we detected a marginally significant correlation that PD patients with higher BDI-II scores tended to present more severe motor symptoms. Olfactory dysfunction seemed to be less relevant to BDI-II scores. Based on our findings, we speculate that depression may be more closely related to brainstem nuclei than to the limbic pathway.