String replace using concatenated strings from various columns

I’d like to remove a substring in a column via update statement. The substring to replace consists of multiple strings from other different columns but in strict order.

The specification says:

replace(string text, from text, to text)

Example:

field1                     |field2       |field3      |field4 
---------------------------+-------------+------------+---------------
(varchar)bla abla 123 ooops|(varchar)abla|(varchar)123|(varchar) ooops

The substring to replace consists of strings in field2, field3, field4 and blanks between field2 and field3, field3 and field4. In the given example: “abla 123 ooops”.

So, after performing the update statement field1 should only contain the string “bla”.

Note: I want to perform this on all rows and not to a particular one.