Biological invasion is regarded as one of the greatest environmental problems facilitated by globalization. Some hypotheses about the invasive mechanisms of alien invasive plants consider the plant–herbivore interaction and the role of plant defense in this interaction. For example, the “Shift Defense Hypothesis” (SDH) argues that introduced plants evolve higher levels of qualitative defense chemicals and decreased levels of quantitative defense, as they are released of the selective pressures from specialist herbivores but still face attack from generalists. Common groundsel (Senecio vulgaris), originating from Europe, is a cosmopolitan invasive plant in temperate regions. As in other Senecio species, S. vulgaris contains pyrrolizidine alkaloids (PAs) as characteristic qualitative defense compounds. In this study, S. vulgaris plants originating from native and invasive ranges (Europe and China, respectively) were grown under identical conditions and harvested upon flowering. PA composition and concentration in shoot and root samples were determined using Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS). We investigated the differences between native and invasive S. vulgaris populations with regard to quantitative and qualitative variation of PAs. We identified 20 PAs, among which senecionine, senecionine N-oxide, integerrimine N-oxide and seneciphylline N-oxide were dominant in the roots. In the shoots, in addition to the 4 PAs dominant in roots, retrorsine N-oxide, spartioidine N-oxide and 2 non-identified PAs were also prevalent. The roots possessed a lower PA diversity but a higher total PA concentration than the shoots. Most individual PAs as well as the total PA concentration were strongly positively correlated between the roots and shoots. Both native and invasive S. vulgaris populations shared the pattern described above. However, there was a slight trend indicating lower PA diversity and lower total PA concentration in invasive S. vulgaris populations than native populations, which is not consistent with the prediction of SDH.