This is what Maciamo seems to advocate. According to him:
(...)
"Nowadays G2a is found mostly in mountainous regions of Europe, for example, in the Apennine mountains (15 to 25%) and Sardinia (12%) in Italy, Cantabria (10%) and Asturias (8%) in northern Spain, Austria (8%), Auvergne (8%) and Provence (7%) in south-east France, Switzerland (7.5%), the mountainous parts of Bohemia (5 to 10%), Romania (6.5%) and Greece (6.5%). It may be because Caucasian farmers sought hilly terrain similar to their original homeland, perhaps well suited to the raising of goats. But it is more likely that G2a farmers escaped from Bronze-Age invaders, such as the Indo-Europeans and found shelter into the mountains. For example, G2a3a (M406) is found at relatively high frequencies in the southern Balkans, the Apennines and the Alps, in contrast with G2a3b (L141.1), which is found everywhere in Europe."
(...)
"Contrarily to other branches of G2a, which are more prevalent in mountainous areas, G2a3b (L141.1), and particularly the G2a3b1 (P303) subclade, is found uniformly throughout Europe, even in Scandinavia and Russia. More importantly, G2a3b and its subclades are also found in eastern Anatolia, the Caucasus, Central Asia and throughout India, especially among the upper castes, who represent the descendants of the Bronze Age Indo-European invaders. The combined presence of G2a3b1 across Europe and India is a very strong argument in favour of an Indo-European origin. The coalescence age of G2a3b1 also matches the time of the Indo-European expansion during the Bronze Age.The homeland of R1b1a (P297) and Pre-Proto-Indo-European speakers is presumed to have been situated in eastern Anatolia and/or the North Caucasus. The Caucasus itself is a hotspot of haplogroup G. Therefore, it is entirely conceivable that a minority of Caucasian men belonging to haplogroup G (and perhaps also J2b) integrated the R1b community that crossed the Caucasus and established themselves on the northern and eastern shores of the Black Sea sometime between 7,000 and 4,500 BCE.An alternative theory is that G2a3 (L30) came from Anatolia to eastern and Central Europe during the Neolithic (a fact proven by ancient DNA test). Once in Southeast Europe it split in two branches: G2a3a, who followed the Danube to Central Europe (LBK), and G2a3b, who migrated east to the Pontic Steppe and brought agriculture to the region. G2a3b would have mixed with the indigenous R1a people, then with R1b newcomers during the Chalcolithic and Bronze Age. By the time the Proto-Indo-Europeans started their massive expansion, G2a3b men (who apparently belonged overwhelmingly to G2a3b1 and its subclades) would have joined R1b-M269/L23 in the invasion of Old Europe from 4200 BCE (=> see R1b history). G2a3a would have been among the conquered populations of Old Europe, seeking refuge in mountainous areas."
(...)
"By the Iron Age, the G2a population in most of Europe had been decimated by the Indo-European invasions, followed by Celtic warfare. G2a sought refuge from the invaders in the mountains, and like today, reached maximum frequencies in Italy (Apennines, Sardinia) and in the Alps.The ancient Latins and Romans descend from the Italic tribes who invaded the Italian peninsula from 1200 BCE. They seem to have belonged primarily to haplogroup R1b-U152 (=> see Genetics of the Italian people), but to have carried a substantial minority of G2a3b (L141.1) lineages, especially the U1 and L497 subclades. The Latin homeland in central Italy is one of the hotspots for haplogroup G2a in Europe today. The high level in G2a in the Latium might be due to the dual presence of Indo-European G2a3b and of earlier Neolithic lineages who descended from the Apennines to live in Rome after being absorbed by the Roman civilisation.If the ancient Romans and other Romanised peoples from the Italian peninsula had any genetic impact on other parts of the Roman Empire (as they should have), they certainly contributed to a moderate increase of G2a lineages (in addition to R1b-U152 and J2) within the borders of the empire. Indeed, the frequency of haplogroup G decreases with the distance from the boundaries of the empire. Haplogroup G is extremely rare Nordic and Baltic countries nowadays, despite the fact that agriculture reached those regions around the same time as Britain or Ireland. Another reason could be that the forested lowlands of northern Germany, Poland and the Baltic were too poor in metals and did not have attract as many Bronze-Age workers from the Caucasus (=> see Metal-mining and stockbreeding explain R1b dominance in Atlantic fringe). Northeast Europe also has a relatively low percentage of haplogroup R1b, which further reinforces the hypothesis that the two haplogroups spread together during the Bronze Age."
The austrian one would not be the subgroup G-L42? Well, any way, G-L497 would be coming from caucasian G-P303, which are present in almost 90% of shapsugs. It reached even India. This is consistent with a more recent expansion of some G2a branches.