The papers, published in The New England Journal of Medicine (NEJM), are based on experiments using blood samples from vaccine trial participants.
But experts believe most coronavirus vaccines should still prevent severe disease.
Scientists from Pfizer and the University of Texas Medical Branch (UTMB) created an engineered virus that contained the same mutations as the South African coronavirus variant, known as B.1.351.
The lab-made virus was then tested against blood taken from 15 people who had been given the BioNTech/Pfizer vaccine.
Results showed there were two-thirds fewer neutralising antibodies against the virus with the key South African mutations, when compared with the most common version of the virus circulating in the US.
For the Moderna study, the team also looked at blood samples of people that had been given their vaccine.
Results showed a six-fold decrease in antibody response against the variant.
Previous research has shown both vaccines were likely to be effective against B.1.1.7, which first emerged in the UK and is also known as the Kent variant.
Preliminary findings from another piece of research has also shown the Oxford/AstraZeneca vaccine to provide "minimal protection" against mild to moderate coronavirus cases caused by the South African variant.
However, experts say it is not possible to say how much effectiveness will be impaired as the research did not include clinical trials.
Lawrence Young, professor of molecular oncology at Warwick Medical School, said: “Two preliminary reports confirm that the South African variant of Sars-CoV-2 is more resistant to antibodies induced by mRNA vaccines.
He added: "The implications of these studies for the efficacy of these vaccines against the South African variant remain to be determined."
Both the Pfizer/BioNTech and Moderna jabs use mRNA vaccines.
Dr Peter English, consultant in communicable disease control and former editor of Vaccines in Practice Magazine, said: "While these findings might, possibly, suggest that the vaccine will be less effective against the South Africa B.1.351 variant, this remains only a possibility - it is far from proven.
"There seems to be little suggestion that the South Africa variant has a particular competitive advantage over other variants - at least in people who haven’t been vaccinated.
"This means that the variant seems unlikely to be a predominant strain in the near future – so, even if it (a vaccine) is less effective against the variant, there won’t be many people exposed to the new variant, and the vaccine will continue to be highly effective in the medium term.
"I think it relatively unlikely that the B.1.351 variant will become a significant problem, but, if it does, by the time it does so, we will almost certainly have vaccines available which are tailored to and will be highly effective against the variant."