We present a search for CO emission in a sample of ten type-2 quasar host galaxies with redshifts of z \approx 0.1-0.4 .
We detect CO ( J=1–0 ) line emission with \geq 5 \sigma in the velocity integrated intensity maps of five sources .
A sixth source shows a tentative detection at the \sim 4.5 \sigma level of its CO ( J=1–0 ) line emission .
The CO emission of all six sources is spatially coincident with the position at optical , infrared or radio wavelengths .
The spectroscopic redshifts derived from the CO ( J=1–0 ) line are very close to the photometric ones for all five detections except for the tentative detection for which we find a much larger discrepancy .
We derive gas masses of \sim ( 2-16 ) \times 10 ^ { 9 } M _ { \odot } for the CO emission in the six detected sources , while we constrain the gas masses to upper limits of M _ { gas } \leq 8 \times 10 ^ { 9 } M _ { \odot } for the four non-detections .
These values are of the order or slightly lower than those derived for type-1 quasars .
The line profiles of the CO ( J=1–0 ) emission are rather narrow ( \lesssim 300 km s ^ { -1 } ) and single peaked , unveiling no typical signatures for current or recent merger activity , and are comparable to that of type-1 quasars .
However , at least one of the observed sources shows a tidal-tail like emission in the optical that is indicative for an on-going or past merging event .

We also address the problem of detecting spurious \sim 5 \sigma emission peaks within the field of view .