Referee 1's review:

This manuscript by Neres et al. describes a potential experimental model to study pregnancy-associated malaria (PAM) in P. berghei ANKA-infected mice. For this propose, authors show several features shared between the presented model and the human severe malaria in pregnancy due to P. falciparum infection.

General Comments

In this paper there are some interesting results. It is clear that pregnant BALB/c mice present pathological features in common with human PAM; such as: post-natal growth impairment, intra-uterine growth retardation, reduction of placental vascular spaces and placental tissue thickening and disorganization. Nevertheless, some points should be clarified:

Point 1:
It is shown by morphometric analyses a reduction in the placental vascular space. Based on these analyses, authors assumed that this vascular space reduction could lead to insufficient blood flow, and in turn, to low oxygen availability. In this sense, I suggest a quantitative analysis by Real-Time PCR, using the cDNA already prepared to chemokine quantification, to check levels of Hypoxia-inducible factor-1alfa (HIF-1alfa).

Point 2:
Despite it is stated that CSA and HA are involved in this mouse experimental model of PAM, there are some points that deserve particular attention:
(i) In the static cytoadhesion assays, it was used placental tissue fixed in formalin. However, I am concerned about the fact that this tissue preparation process could damage placental receptors, including CSA. It would be preferable to use tissue from cryo-preserved placentas prepared according to Avril et al. 2004 (Microbes and Infection 6: 249-255). Alternatively, placental tissue can be cryo-preserved in N2-frozen hexane, then fixed in Tissue-tek solution prior to the cutting process.
(ii) The amount of soluble CSA (2.000 ug) used in the experiments (Figure 8) is highly above the levels used in cytoadhesion assays with P. falciparum-CSA parasites.
(iii) It is questionable why CaseABC treatment did not reduce parasite adhesion to the same levels than soluble CSA treatment. Indeed, according to Figure 8, previous incubation with CaseABC inhibited 30% vs. 50% with soluble CSA at 2.000 ug. It seems that part of the inhibitory effect observed is due to alosteric impairment and brings to light concerns about the specificity of the inhibition assays in the presence of soluble CSA. Authors should provide dose-dependent experiments using different concentrations of CSA and HA. Moreover, these assays should be performed in the presence of CSB, as a negative control for CSA.
(iv) If CSA or HA receptors are involved in P.berghei adhesion to the placental tissue, and this adhesion is the key point to the inflammatory process and, in turn, to all placental abnormalities and their consequences to fetal development; alternatively the authors should compare fetus weight (G18) between a group of mice treated with CSA or HA vs. a non-treated mice.

Minor points
1- Figure 6 is missing
2- In the title should be mentioned P. berghei ANKA, instead of P. berghei.

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.